ArticlePDF Available

Fuzzy Support Tensor Product Adaptive Image Classification for the Internet of Things


Abstract and Figures

Computer vision is one of the hottest research directions in artificial intelligence at present, and its research goal is to give computers the ability to perceive and cognize their surroundings from a single image. Image recognition is an important research direction in the field of computer vision, which has important research significance and application value in industrial applications such as video surveillance, biometric identification, unmanned vehicles, human-computer interaction, and medical image recognition. In this article, we propose an end-to-end, pixel-to-pixel IoT-oriented fuzzy support tensor product adaptive image classification method. Considering the problem that traditional support tensor product classification methods are difficult to directly produce pixel-to-pixel classification results, the research is based on the idea of inverse convolution network design, which directly outputs dense pixel-by-pixel classification results for images to be classified of arbitrary size to achieve true end-to-end and pixel-to-pixel high-score image classification and improve the efficiency of support tensor product models for high-score image classification on a pixel-by-pixel basis. Moreover, considering that network supervised classification training using deep learning requires a large amount of labeled data as true values and obtaining a large number of labeled data sources is a difficult problem in the field of image classification, this article proposes using a large amount of unlabeled high-resolution remote sensing images for learning generic structured features through unsupervised to assist the labeled high-resolution remote sensing images for better-supervised feature extraction and classification training. By finding a balance between generic structural feature learning of images and differentiated feature learning related to the target class, the dependence of supervised classification on the number of labeled samples is reduced, and the network robustness of the support tensor product algorithm is improved under a small number of labeled training samples.
This content is subject to copyright.
Research Article
Fuzzy Support Tensor Product Adaptive Image Classification for
the Internet of Things
Zhongrong Shi ,
Yun Ma ,
and Maosheng Fu
Faculty of Electronic and Information Engineering, West Anhui University, Lu’An, Anhui 237012, China
Anhui Yongcheng Electronic and Mechanical Technology Co., Ltd.,, Lu’An, Anhui 237000, China
Faculty of Electrical and Opto-Electronic Engineering, West Anhui University, Lu’An, Anhui 237012, China
Correspondence should be addressed to Yun Ma;
Received 12 November 2021; Revised 18 January 2022; Accepted 21 January 2022; Published 22 February 2022
Academic Editor: Akshi Kumar
Copyright ©2022 Zhongrong Shi et al. is is an open access article distributed under the Creative Commons Attribution License,
which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
Computer vision is one of the hottest research directions in artificial intelligence at present, and its research goal is to give
computers the ability to perceive and cognize their surroundings from a single image. Image recognition is an important research
direction in the field of computer vision, which has important research significance and application value in industrial ap-
plications such as video surveillance, biometric identification, unmanned vehicles, human-computer interaction, and medical
image recognition. In this article, we propose an end-to-end, pixel-to-pixel IoT-oriented fuzzy support tensor product adaptive
image classification method. Considering the problem that traditional support tensor product classification methods are difficult
to directly produce pixel-to-pixel classification results, the research is based on the idea of inverse convolution network design,
which directly outputs dense pixel-by-pixel classification results for images to be classified of arbitrary size to achieve true end-to-
end and pixel-to-pixel high-score image classification and improve the efficiency of support tensor product models for high-score
image classification on a pixel-by-pixel basis. Moreover, considering that network supervised classification training using deep
learning requires a large amount of labeled data as true values and obtaining a large number of labeled data sources is a difficult
problem in the field of image classification, this article proposes using a large amount of unlabeled high-resolution remote sensing
images for learning generic structured features through unsupervised to assist the labeled high-resolution remote sensing images
for better-supervised feature extraction and classification training. By finding a balance between generic structural feature learning
of images and differentiated feature learning related to the target class, the dependence of supervised classification on the number
of labeled samples is reduced, and the network robustness of the support tensor product algorithm is improved under a small
number of labeled training samples.
1. Introduction
Image recognition technology is an important research
branch in the field of computer vision, which aims to identify
various potential objects in images using computers to
preprocess, extract features, analyze and understand them.
Traditional image recognition models can be divided into
two parts: excellent feature extraction methods are robust in
various complex environments, while classifiers mainly
consist of some shallow machine learning algorithms for
predicting the classes to which the features obtained by the
extractor belong. For example, in the field of face recog-
nition, researchers can improve the accuracy of a large
number of face image samples by simply feeding them to a
model, which can then be trained over some time by deep
learning models by more than a dozen percentage points
over traditional models [1]. Excellent feature extraction
methods are robust in various complex environments, while
classifiers mainly consist of some shallow machine learning
algorithms for predicting the classes to which the features
obtained by the extractor belong. Target detection, one of the
most challenging research hotspots in computer vision, is
also a fundamental technique for solving more complex and
advanced vision tasks such as semantic segmentation, target
tracking, image description, and scene recognition. Deep
learning techniques, which have emerged in recent years, are
Computational Intelligence and Neuroscience
Volume 2022, Article ID 3532605, 11 pages
a powerful method for learning feature representations
directly from data and have already brought breakthroughs
in the field of target detection. Currently, the most popular
image recognition techniques include image classification,
target detection, target tracking, and semantic segmentation,
and a huge amount of problems in computer vision can be
solved perfectly using these four techniques, either directly
or indirectly. ese techniques assist computers in
extracting, analyzing, and understanding useful information
from a single or a series of images, and the trend is for deep
learning models to gradually replace traditional latent rec-
ognition models. However, while various popular algo-
rithms or models embody excellent performance, there are
still some general problems and challenges, such as the
insufficient number of samples, poor target viewability, and
slow convergence of models to train.
With the rapid development of artificial intelligence
techniques and computer performance, image recognition
based on machine learning algorithms has rapidly expanded
from targeted application scenarios to become a standard
scientific tool and applied to the full range of natural science
and technology applications. In this article, we propose a
novel adaptive image classification method, the particle
swarm optimized fuzzy support tensor machine [2]. e
method firstly calculates the fuzzy affiliation of each sample
by fuzzy affiliation function to reduce the influence of noise
points on the classification results; secondly, it uses particle
swarm algorithm to perform parameter search for fuzzy
support tensor machine; although it is more concise and easy
to operate compared with other commonly used parameter
optimization algorithms such as genetic algorithm and least
squares method, the particle swarm algorithm still has the
disadvantage of easily falling into local optimum in this
article[3]. To solve this problem, the particle swarm algo-
rithm is improved: firstly, the inertia weights are introduced
into the particle swarm algorithm, which decreases non-
linearly with the number of iterations to improve the al-
gorithm’s optimality seeking ability, and secondly, the
simulated image recognition algorithm is used to make the
particles in the particle swarm algorithm forcefully jump out
of the local optimum trap with a certain probability. e
improved particle swarm algorithm greatly improves the
efficiency of the optimization search and overcomes the
blindness of parameter selection in the traditional classifi-
cation model.
2. Related Work
Unlike pixel-based classification methods, object-oriented
classification methods use the segmented objects or regions
of an image as the minimum unit of analysis to compensate
for the lack of contextual or spatial relationships in pixel-
based classification methods. Object-oriented classification
methods usually use segmentation before classification and
obtain the classification results of the image by extracting
features from the segmented results and then training the
In the literature [4], a nonparametric Bayesian hierar-
chical model is proposed for high-resolution remote sensing
image classification using a combination of object-oriented
oversegmentation and hierarchical Dirichlet process model
(HDP) and Indian buffet process (IBP), which solves a series
of problems such as the traditional probabilistic topic model
and ignores spatial information, and the number of topics
has to be predetermined; in the literature [5], a remote
sensing image is proposed for object-oriented Markov re-
gion penalty method for remote sensing images, using mean
shift algorithm for image segmentation and establishing
weighted region neighborhood map based on region size and
neighboring region connection strength, while using region
size and neighborhood strength features as penalty terms to
calculate potential functions, and using maximum posterior
probability to iteratively update joint probability distribu-
tion and likelihood function to obtain the final semantic
segmentation results. is approach can better weigh the
interactions between neighbors and obtain more macro-
scopic texture features; in the literature [6], after superpixel
segmentation using the simple linear iterative clustering
(SLIC) method, visual features are extracted and trained on
the mixture components generated by the Dirichlet process
mixture model through a multiple conditional random field
model to obtain intermediate labels corresponding to visual
features in the new feature space. Using the intermediate
labels, further pixel semantic analysis is performed to es-
tablish the connection between low-level features and high-
level semantics based on the spatial relationships taking into
account the objects; literature [7] uses the watershed algo-
rithm to record the process of gradual merging of over-
segmented regions by region similarity comparison after
oversegmenting the image using a binary segmentation tree,
on which the ascending trajectory between the leaf nodes to
the root node is the region evolution process obtained
computationally, and the search for salient region content is
achieved by finding the maximum value through first-order
derivation of the evolution value, which improves the
previous complex tree construction model and thus achieves
image classification and target recognition more simply and
efficiently. In the literature [8], a region of interest detection
algorithm based on the MFF (Multiscale Feature Fusion)
algorithm is proposed, which detects the region of interest in
remote sensing images accurately and quickly by performing
grayscale saliency analysis based on multiscale spectral re-
siduals and directional saliency analysis based on integer
wavelet transform on remote sensing images. In the liter-
ature [9], by performing Saliency Analysis of Cooccurrence
Histogram (SACH) based on cooccurrence histogram for
high-resolution remote sensing images and using a saliency
enhancement method based on moving K-means clustering,
clear region boundaries are established for the region of
interest, while improving the immunity of the algorithm to
noise. To reduce the computational complexity of region of
interest detection for remote sensing images, the literature
[10] proposes to achieve fast and efficient region of interest
detection by segmenting high-resolution remote sensing
images into superpixels and generating superpixel-level
saliency maps using structural tensor and background
contrast, and finally by superpixel-to-pixel-based saliency
analysis. To extract high-quality regions of interest with clear
2Computational Intelligence and Neuroscience
boundaries and no background interference from remote
sensing images, a GLSA (Global and Local Saliency Analysis)
algorithm based on global and local saliency analysis is
proposed in the literature [11] for extracting residential
regions in high-resolution remote sensing images. In ad-
dition, for common features of interest in high-resolution
remote sensing images, such as residential areas, airports,
aircraft, and ships, a detection algorithm based on joint
multi-image saliency (JMS) is proposed in the literature [12],
which processes multiple multispectral remote sensing
images with similar spatial structure and spectral details by
jointly using the correlation information between this set of
remote sensing images to simultaneously detect the features
of interest in this set of multispectral remote sensing images.
e literature [13] proposes a region of interest detection
algorithm based on superpixel segmentation and statistical
significance analysis, which detects the region of interest in
remote sensing images accurately based on the final gen-
erated significance map by fusing the statistical significance
feature map based on histogram statistics and the infor-
mation significance feature map based on information en-
tropy analysis. e literature [14] proposes that candidate
regions containing feature targets can be predicted by su-
pervised learning models constructed from various salient
features. en a discriminative dictionary learning classifier
based on sparse coding representation can be applied to the
target candidate regions to detect feature targets in the scene,
which greatly reduces the computational cost of traditional
search strategies. For the airport detection problem of
panchromatic remote sensing images, the literature [15]
proposes to use a graph-based visual saliency model to locate
the salient regions in the scene and obtain a top-down sa-
liency map by making full use of the geometric prior
knowledge of the airport runway, and finally by combining
these two saliency maps, to predict the location of the airport
more accurately. In the literature [16], a two-layer visual
saliency analysis model is proposed to extract candidate
regions of airports and aircraft, and a bag-of-words model
based on dense SIFT features and Hu moment features are
used to characterize the invariant features of airports and
aircraft, and finally, the airport and aircraft targets in remote
sensing images are accurately detected by support tensor
3. Fuzzy Support Tensor Machine Adaptive
Image Classification for the
Internet of Things
3.1. Fuzzy Support Tensor Machine eory. Support tensor
machine based on statistical learning theory relies on its
excellent learning ability and powerful generalization ability
to have better results in dealing with problems with a small
amount of sample data and nonlinear relationship between
samples. However, there are still shortcomings in the sup-
port tensor machine model. e support tensor machine
works by determining the support tensor with a small
number of training samples and then finding a classification
surface that can divide the samples and then classify the test
samples. If there are incorrect or biased samples in the
training samples to determine the support tensor, the
support tensor machine cannot exclude these samples be-
cause they all have the same reliability for the test samples, so
it is easy to be misled by these incorrect or biased test
samples and establish the wrong optimal hyperplane,
resulting in a decrease in the classification accuracy of the
model. e biggest difference between standard fuzzy
support tensor machine and support tensor machine is that
the former has one more dimension than the latter, i.e., fuzzy
affiliation. Hence, the support tensor selected by the fuzzy
support tensor machine is not equivalent to that selected by
the support tensor machine. To address this problem, fuzzy
support tensor machines introduce the concept of a fuzzy
affiliation function [17]. Each training sample is assigned a
corresponding fuzzy affiliation according to its influence on
the prediction result, with smaller affiliations for incorrect or
biased samples and larger affiliations for correct samples, by
which the problem that traditional support tensor machines
are easily misled by isolated points is solved and the noise
immunity of the model is improved.
As the complexity of the research problem increased, the
degree of state in the problem was vague and could not be
accurately described by traditional exact mathematics, and
fuzzy mathematics was created to have a reasonable de-
scription of the degree of state of certain factors in the
problem. e discipline was introduced in the 1950s and was
mainly used to study some fuzzy problems. With the rapid
development of artificial intelligence technology, fuzzy
mathematics has been combined with various intelligent
algorithms and is widely used in various fields. A fuzzy set is
a basic concept of fuzzy mathematics. In the traditional
notion of set, for an individual uand a set A, the relationship
between them is that ueither belongs to A or does not belong
to A. ese two results cannot hold simultaneously [18]. e
relationship between an individual and a set, if expressed by
a mathematical expression, should be
Cij 􏽘
e eigenfunction cannot explicitly define data as be-
longing to a certain state or not belonging to a certain state.
In the 1960s, researchers used feature functions to express an
increasing number of classical sets by representing each data
in the set as a fuzzy number, such that the domain of values
of fuzzy sets was extended from the set of integers {0,1} to the
set of real numbers [0,1]. Since the value of the fuzzy af-
filiation reflects the training points, for a certain class of
defined affiliations, and the parameter jis a measure of the
extent to which the support tensor machine misclassifies the
samples, combining the two becomes a measure of how
correctly the support tensor machine classifies data with
different affiliations. Each training sample is assigned a
corresponding fuzzy affiliation according to its influence on
the prediction result, with a smaller affiliation for incorrect
or biased samples and a larger affiliation for correct samples.
By this approach, the problem that traditional support
tensor machines are easily misled by isolated points is solved,
Computational Intelligence and Neuroscience 3
and the noise immunity of the model is improved. Trans-
ferring the processed data as input to the prediction model,
the process of finding the optimal hyperplane for the
classification model can be expressed mathematically as a
quadratic program if the data transferred is linearly divisible.
Eall 􏽘
Econtest ×x+Ee ×x+εfree ×x2×d(1/2)
􏼐 􏼑
4Ee×x+εdecay ×x2×d2
􏼐 􏼑.
Minimizing αin the objective function yields the qua-
dratic counterpart of the pairwise plan:
I(X;Y) � K(x, y) · 􏽐n
e major difference between a standard fuzzy support
tensor machine and a support tensor machine is that the
former has one more dimension, i.e., fuzzy affiliation, than
the latter, so the support tensor selected by fuzzy support
tensor machine is not equivalent to that selected by support
tensor machine. If the relationship between the factors in the
problem to be solved is nonlinear and a kernel function is
introduced in the solution process, then the classification
problem can be expressed in mathematical form as follows:
H(x) � φ􏽘
[p(x, y) · ln p(x, y) + Ax +Cy] + λ.(4)
One way to represent the above diffusion tensor is to use
the covariance matrix of the Gaussian distribution, which
mainly describes the diffusion process of water molecules in
the tissue. e statistical scatter of the two distributions is
used. In this case, tensor comparisons are made by mea-
suring the Kullback–Leibler distance between the probability
distributions afterward. e symmetric version is called J
scatter, which was proposed by Wang and Vemuri and used
for tensor distance measurements. It is shown as follows:
M(x) � φ􏽘
p2(x) · ln p(x)
􏽨 􏽩 +Ax. (5)
Although Support Tensor Machines (STM) solve the
overfitting problem in traditional SVMs, the rank-weight
tensor is weakly expressive and this translation leads to
poorer classification accuracy. e rank-weight tensor of
STM is generalized to Tucker decomposition and CP forms
to obtain stronger model expressiveness. However, the CP
rank-decomposition causes an exponential increase in the
number of parameters in the Tucker form, which suffers
from dimensional catastrophe.
3.2. Fuzzy Support Tensor Machine Adaptive Image Classifi-
cation for the Internet of ings. Classification algorithm
design has been a hot topic in the field of machine learning,
pattern recognition, and computer vision. One of the most
representative and successful classification algorithms is the
Support Vector Machine (SVM), which has been highly
successful in pattern classification by minimizing the
Vapnik–Chervonenkis dimensional and structural risk.
However, standard SVM models are based on vector inputs
and cannot directly deal with matrices or higher-dimen-
sional data structures, i.e., tensors, which are very common
in real life. As in Figure 1, a grayscale image is a two-di-
mensional matrix with height and width, which is a second-
order tensor, while a multispectrum has multiple spectral
bands and is a third-order tensor. When high-dimensional
data are fed into the SVM, a common approach is reshaping
each sample into a vector. A tensor can be seen as an ex-
tension of a matrix, which in traditional signal research can
be considered as an array of different dimensions depending
on the object of study. However, when the training data
sample size is relatively small concerning the dimensionality
of the feature vectors, this can result in overfitting and lead to
unsatisfactory classification performance.
e traditional image recognition model consists of two
parts: feature extractor and classifier. Feature extraction
methods can be classified into texture features, shape fea-
tures, bag-of-words model, sparse coding, local coding, and
Fisher vectors. ese feature extraction extracts feature from
the image and then a set of numbers or symbols are used to
represent certain characteristics of the depicted object in the
image and finally, these features are recognized with the help
of other machine learning methods. e classical recognition
methods (classifiers) are support vector machines, decision
trees, adaptive enhancement, plain Bayes, and some heu-
ristic arithmetic [19]. ese classical feature representation
methods have a common feature that they all require a very
specialized knowledgeable researcher to carefully design the
model; however, this feature makes the model deficient in
two ways: firstly, the researcher needs to spend a lot of effort
to design different features for different recognition tasks;
secondly, the practical application requires repeated vali-
dation and parameter tuning of the model, which is very
costly to optimize.
Wavelet theory is an extended version of Fourier
transform theory in which a signal is decomposed into
wavelets and projected onto a set of wavelet functions. is
differs from the Fourier transform, which decomposes the
signal into sine and cosine components. Wavelet transform
theory is popular in image processing, where it decomposes
the input image into a set of images with various resolutions
while reducing redundancy in the image representation [20].
e parent signal components are decomposed in an ex-
tended signal variant or a shifted wavelet. Two basic
properties must be satisfied for a wavelet to be considered a
wavelet. e information flow used for a single-level or one-
level 2D image decomposition scheme is illustrated in
Figure 2.
e inverse 2D wavelet transform used to reconstruct the
image involves column upsampling and filtering for each
subimage using low-pass and high-pass filters. e initial
source image is constructed using the low-pass filter Land
the high-pass filter of the resulting image for row upsam-
pling and filtering and the summation of all matrices. By
examining the saliency type focusing on bottom-up, the
images can be classified into two categories, spatial domain
models and transform domain models, depending on
4Computational Intelligence and Neuroscience
whether they are transformed in the frequency domain [21].
e so-called spatial domain saliency models process the
image directly in the spatial domain and thus detect the
salient targets or regions of interest in the image. erefore,
in the design of saliency models, regions in the image scene
with unique color features or pattern features should have
high saliency, while homogeneous regions in the scene
should have low saliency. Features that frequently appear in
the image scene should be suppressed. Salient pixels in an
image should be clustered together rather than scattered
throughout the image. erefore, the Euclidean distance in
spatial location of image blocks containing contextual
information about each pixel in the image is also important,
because the distribution of image blocks in the background
region is either far or near in space, while the distribution of
image blocks in the salient target region tends to be clustered
together in space. e saliency detection results can be
further enhanced by incorporating the central prior
knowledge of the salient regions. e saliency of pixel I in an
image at a single scale can be defined as follows:
k(t) � 􏽘
􏽨 􏽩2ηi,j(t)
􏽨 􏽩β.(6)
image data
The original gradient
The proccessed
gradient image histogram The mapped histogram
thresholdThe original threshold
Edge image
Sobel operator
NMS proccessing Mapping fuction
OTSU method
Inverse mapping
Calculation double
Binarization by
double threshold
Figure 1: Picture and tensor representation.
Risk evaluation
Many risks
Dicult to highlight Focus of prevention Risks are included Risk evaluation index
S1 S3S2
S4 S6S5
Risk evaluation
Practicality and
Index system
Low frequency signal Horizontal signal
Taking into account
the focus and
While avoiding the
mutual inuence
Figure 2: Wavelet decomposition of the two-dimensional image.
Computational Intelligence and Neuroscience 5
Furthermore, considering that image blocks in the
background region are similar at multiple scales, in
contrast, image blocks in the saliency region may be
similar at only a few scales but not at all scales. erefore,
multiple scales can be used to further reduce the saliency
of background pixels and enhance the contrast between
salient and nonsalient regions. Unlike the spatial domain
saliency model, the transform domain-based saliency
model requires first transforming the image from the
spatial domain to the frequency domain, then processing
and analyzing the image in the frequency domain, and
finally obtaining the final saliency detection results by
transforming the analysis results in the frequency domain
back to the spatial domain. For a two-dimensional signal
like an image, by performing the Fourier transform on it,
the resulting image amplitude spectrum clarifies the
percentage of each sinusoidal component, while the phase
spectrum of the image gives the position of each sinusoidal
component in the graph. In the reconstruction of the
image in the Fourier transform domain, the positions
located in the horizontal or vertical directions with weak
periodicity or homogeneity correspond to the positions of
the candidate targets in the image, and thus it is known
that the saliency information of the image is implicit in the
phase spectrum of the image. erefore, the saliency
detection result of the image can be obtained by extracting
the phase spectrum information of the image. e initial
saliency analysis results are smoothed by a two-dimen-
sional Gaussian filtering function g(x, y)to obtain a vi-
sually superior final saliency map as follows:
ej� −k􏽘
piln 1
From the above analysis, it can be seen that the PFT
transform domain saliency detection model has the ad-
vantages of simple and easy algorithm and fast operation,
thus giving fast saliency detection results for a given image,
but the disadvantage is that the local saliency features of the
image are not considered, and it lacks suitable biological
psychological support and explanation. After each iteration
step, the optimization progress of the solution needs to be
measured by a predefined criterion that determines whether
the current state is the best fit or not. Among the saliency
analysis models for natural images, the center prior and the
boundary prior are the two most widely used prior
knowledge, which achieves good detection results in the
saliency detection of natural images, thanks to the imaging
mechanism of natural images, where the salient targets of
natural images are usually at the location of the image center,
while the boundaries of natural images usually do not have a
distribution of salient targets.
Using superpixel segmentation methods, an image is
segmented into many superpixels. Each superpixel contains
a large number of spatially close neighboring samples that
have a similar texture, color, luminance, and other char-
acteristics. Compared to pixel-based hyperspectral image
classification methods, the superpixel-based classification
methods demonstrate good regional consistency.
e superpixel segmentation process is shown in
Figure 3.
To alleviate pseudoboundaries that cause misclassifica-
tion, we propose a new nonlocal decision-based region
delineation method. In hyperspectral images, we usually
consider that the samples in local regions belong to the same
class, which is local information. However, nonlocal in-
formation is also very critical in hyperspectral images. is is
because samples of the same class may also be located in
different regions of the image. In nonlocal decision making,
pixel pair similarity is extended to superpixel pair similarity,
taking into account the structural information of the current
samples. For those samples that are judged to be in het-
erogeneous regions by the local decision, the similarity
between the current sample and the filter neighborhood
samples is calculated. e current sample is represented by a
global search for all similar samples. en, this similarity is
compared with the calculated adaptive threshold. If the
similarity of all neighborhood pixels all is greater than the
threshold, the current sample is judged to be in the ho-
mogeneous region and vice versa. is model includes two
stages: the first stage is to enhance the input image, then
input the residual network to carry on the supervised
contrast learning, and get the pretraining model; the second
stage is to fix the parameters of the pretraining model and
the fuzzy support tensor machine is trained to get the
prediction label. In the first stage, in order to enhance the
discriminative ability of feature extraction, local informa-
tion, nonlocal information, and generic structured features
that come from unlabeled high-resolution images are also
introduced, respectively. After an image of arbitrary size is
input to the network as input data, the input data are
convolved by each branch in the subnet separately utilizing
dense convolution operations at different scales to associate
the final extracted results, reduce the dimensionality through
the transition layer, and then use the output data to (1)
obtain the pixel-by-pixel classification results with the
classifier to compare with the reference marker to calculate
the loss; and(2) input them to the next subnetwork. e
above process is repeated until the final objective function is
obtained. e overall loss function is calculated jointly for all
the objective functions and the network is trained by
backpropagation of stochastic gradient descent. To facilitate
training, the weights of the objective functions of all clas-
sifiers are learned in an alternating manner with the learning
of other network parameters.
4. Experimental Verification and Conclusions
We compare the model in this article with Single Task
Learning (STL), Tensor Train Multitasking (TT-MTL), and
Tucker-based multitasking models. For a fair comparison,
the same network architecture is used for all methods. In all
experiments, this article sets the model format to M3 and
N2. Since the model is harder to train when M,Nis larger
because of the presence of a fifth-order tensor. (Perhaps this
can be solved by trying to decompose the larger cores
further.) In this article, we use this relatively lightweight
structure for our experiments. For the choice of rank, the
6Computational Intelligence and Neuroscience
model parameters are extremely large when the rank is
particularly large, so a relatively small rank (3, 4, 5) is used.
e MNIST 10-class classification problem can be converted
to a ten one-vs-all binary classification problem. is con-
version allows the construction of a 10-task classification
problem of the same kind, that is, by performing a softmax
normalization on the ten classifiers before training. In this
experiment, this article focuses on two performance metrics:
one is the average accuracy of the ten binary classification
problems, and the other is the accuracy of classifying a single
digit by performing a softmax on the one-vs-all output of
each task (multiclass classification accuracy). In this article,
the first three convolutional layers are set to be hard-shared
across all MTL models (for common feature extraction), and
then the next FC layer is converted to a different multitask
tensor network model format. In this article, we train with
different sized subsets of the training dataset and test the
model using the same test set (from 10% to 100% of the
training set). As is shown in Figure 4, all MTL methods
outperform STL for both more and less training data. Also,
TT outperforms Tucker when the training data are small,
while the results are reversed when the training data are
large. e CTN proposed in this article outperforms all other
In the whole experimental process, 100,000 training
iterations are set in this article because the experimental
images are few sample data, the choice of iteration pa-
rameters will affect the final model effect, and the accuracy
and loss function of the model training can be analyzed to
see in which interval the network reaches a stable equilib-
rium state so that the network training is in the optimal state,
the accuracy and loss function curves are plotted and dif-
ferent network modules are designed to design the node
embedding network, and the experimental results are shown
in Figure 5. e comparison experiment of SE-ResNet
structure and simple node embedding network access GNN
can be obtained, the red line is the change curve of NSE-
ResNet-EGNN network obtained by node embedding using
SE-ResNet structure, and the blue is the change curve of
simple node embedding network EGNN. is indicates that
the extracted finite number of feature parameters can fully
reflect the transient response of the actual waveform pattern.
From the accuracy change curve, we can see that the two
approaches are almost the same in terms of the convergence
speed in the early stage, but in terms of the final convergence
result, the node embedding using the SE-ResNet network
model can achieve higher accuracy, and in the subsequent
iterations, the trend of the network curve is smoother
compared to the simple node embedding network. From the
loss change curve, we can see that the node embedding using
the SE-ResNet network model makes the loss function fall
faster in the training process and can maintain a smoother
convergence effect compared with the original knot, and
from the final convergence, the node embedding using SE-
ResNet network model will have less loss, and we can see that
by improving the dependency relationship between node
channels, the performance of the node embedding network
can be effectively enhanced by improving the dependencies
between the node channels.
ough the previous experiment, it can be obtained that
improving the node embedding network can increase the
information encoded to the nodes, thus making the edge
features describe the nodes more accurately. e network
design of node update is also available in the GNN-Block
module, so two sets of comparison experiments are set up in
this experiment from the number of Conv-Blocks in the
node update network and the network architecture, re-
spectively. To verify the effect of the number of Conv-Blocks
on the final results, we choose the node update framework as
the comparison experiments, whose experimental results are
shown as the yellow and red curves in Figure 6, respectively.
From the accuracy change curve, we can get that increasing
the Conv-Block of the node update network can improve the
final classification accuracy within a certain range and keep
the same convergence speed in the early stage, but in the final
convergence result, the number of Conv-Block is propor-
tional to the classification accuracy within a certain range
using the node update network. From the loss variation
curve, we can see that as the number of Conv-Blocks of the
nodal update network increases, the loss function decreases
faster due to the more parameters and better robustness of
the network. Regarding network depth, and from the final
MB Supercluster LOB
Standard data
Quant features
Entropy LMS1 LMS2 LDA1 LDA2
Collect the ve sorting lists
Sender 1
Figure 3: Superpixel segmentation process.
Computational Intelligence and Neuroscience 7
improvement results, increasing the complexity of the node
update network can improve the few-sample classification
performance within a certain range.
To visualize the effectiveness of the proposed algorithm,
the distribution of representation coefficients and the cor-
responding normalized reconstruction residuals of the
MFCARC algorithm are given. In our experiments, we se-
lected the Indian Pines dataset for analysis, selected ten
random samples from each feature class to construct the
training dictionary, and randomly selected one sample from
class 6 (grass/trees) of this dataset as the test book and then
analyzed the representation coefficient distribution of this
sample. Since the decomposition calculation process has a
truncation process of redundant columns for the tensor
factor matrix, some errors are inevitable while simplifying
the calculation process. As shown in Figure 7, the distri-
bution of the correlated adaptive representation coefficients
based on different features exhibits the feature that the part
of the representation coefficients with larger weights is
mainly concentrated in the category to which the test sample
belongs. From the minimum representation residual crite-
rion, it is known that the category to which the test sample
belongs should have the smallest normalized residual value
by comparing the reconstruction error of the test sample and
each category of dictionaries. From Figure 7, it can be found
that the correlated adaptive representation models based on
spectral features, DMP features, and LBP features all make
correct feature category determination for the test sample,
but the correlated adaptive representation model based on
Gabor features incorrectly determines the sample as cate-
gory 4 (maize).
Most detection models are less effective in detecting
smaller objects than in detecting larger objects. is is
mainly because, after multiple layers of convolution, small
objects may not retain any information in the feature
mapping at the topmost layer of the model. Increasing the
size of the model input (e.g., from 300 ×300 to 512 ×512) can
help the model improve its performance in detecting small
objects, with a 2.5 percentage point improvement in mAP
for SSD and a 3.2 percentage point improvement in EAO.
e analysis suggests that this is related to the multi-
resolution detection layer proposed in this article, which
gives different detection layers to detect objects of different
sizes, such as the low-resolution detection layer used to
improve the detection rate of small targets. e experimental
results in Figure 8 are from seven smaller objects (bird, boat,
chair, etc.) in PACAL_VOC2007, with XL, L,M,S, and XS in
the horizontal coordinates denoting extra-large, large, me-
dium, small, and very small, respectively, and the vertical
coordinates denote the average detection accuracy of the
model Here the different size pairs are produced by hand
cropping postprocessing formation. From the figure, it can
be seen that EA0 outperforms the base model SSD almost
across the board and the advantage is more pronounced
when the object size is of S and XS level. e analysis suggests
that this is related to EA0’s strategy of using different shapes
Figure 4: e average accuracy of binary classification for different algorithms with the same dataset.
0 200 400 600 800 1000 1200
Number of iterations
Figure 5: Classification accuracy and loss line of the model.
8Computational Intelligence and Neuroscience
3.6 4.8 6.0 7.2 8.4
Sepal Length
Sepal Length
1.62 2.43 3.24 4.05
024680.00 0.81 1.62 2.43
Sepal Width
Petal Length
0.00 0.81 1.62 2.43 0
Petal Width
Petal Width
Figure 6: Classification accuracy and loss lines of the model under different GNN-blocks.
20 30 40 50 60 70 80 90 100 110 120 130 140 150 160 170 180
Normalized residuals
Class index
Figure 7: Normalized residuals for different algorithms.
Computational Intelligence and Neuroscience 9
and numbers of a priori frames at different detection layers
and assigning more detection frames to lower resolution
layers based on the results of the cluster analysis. e best
results achieved by EA0 in all scales of objects also reflect the
lower sensitivity and greater robustness of the model to
bounding box size than SSD.
5. Conclusion
Deep learning has been a great success in the fields of image
recognition, speech recognition, and machine translation.
Among them, support tensor machines have made break-
throughs in image detection, image classification, image
segmentation, face recognition, video tracking, and other
vision-related domains and have achieved great success in
these fields. It is due to the powerful feature extraction
capability of support tensor machines in image classification
that more and more scholars are applying support vector
machines to image classification. Tensor algorithms have
different decompositions in several scientific fields. e
different decomposition methods have their own advantages
and areas of application. In this article, we propose an end-
to-end, pixel-to-pixel IoT-oriented fuzzy support tensor
product adaptive image classification method, introduce the
background of the current topic of image classification and
the significance of the research, as well as the current state of
research on image classification, and draw out the difficulties
faced by existing image classification methods by analyzing
the characteristics of image classification and the advantages
and disadvantages of existing image classification, providing
strong realistic implications.
e accuracy of the prediction of the classification model
established using fuzzy support vector machine and the
selection of parameters of the algorithm have a great rela-
tionship, the more reasonable the selection of parameters,
the higher the accuracy. erefore, in this article, to mini-
mize the influence of parameter selection on the classifi-
cation accuracy of IoT image recognition, nonlocal adaptive
information is introduced, and the improved nonlocal in-
formation is combined with the fuzzy support vector ma-
chine to apply to the image classification research. is new
approach to image classification is also the innovation of this
paper. Experiments show that the model has better per-
formance than standard RBMs in feature extraction and
denoising tasks. Two visible and hidden layers of TRRBM
are represented as matrix product states (MPS) and all
computations can be done on a single kernel. is can
significantly improve the computational complexity of the
learning algorithm.
Data Availability
e data used to support the findings of this study are
available from the corresponding author upon request.
Conflicts of Interest
is author declares that there are no potential conflicts of
interest in this article.
is research was funded by the Outstanding Young Talent
Support Program Project of Anhui Province (gxyq2019071).
[1] S. Smys, A. Basar, and H. Wang, “Hybrid intrusion detection
system for internet of things (IoT),” Journal of ISMAC, vol. 2,
no. 4, pp. 190–199, 2020.
10 25 40 55 70
10 25 40 55 70
10 25 40 55 70
10 25 40 55 70
Comparison of recognition accuracy
Figure 8: Comparison of recognition accuracy of different models for images of different sizes.
10 Computational Intelligence and Neuroscience
[2] R. Rayhana, G. Xiao, and Z. Liu, “Internet of things
empowered smart greenhouse farming,” IEEE Journal of
Radio Frequency Identification, vol. 4, no. 3, pp. 195–211,
[3] X. Wang and S. Gao, “A chaotic image encryption algorithm
based on a counting system and the semi-tensor product,”
Multimedia Tools and Applications, vol. 80, no. 7,
pp. 10301–10322, 2021.
[4] Y. Li, Y. Huang, M. Zhang, and L. Rajabion, “Service selection
mechanisms in the Internet of ings (IoT): a systematic and
comprehensive study,” Cluster Computing, vol. 23, no. 2,
pp. 1163–1183, 2020.
[5] J. Zhang and D. Tao, “Empowering things with intelligence: a
survey of the progress, challenges, and opportunities in ar-
tificial intelligence of things,” IEEE Internet of ings Journal,
vol. 8, no. 10, pp. 7789–7817, 2020.
[6] M. Ruta, F. Scioscia, and G. Loseto, “Machine learning in the
Internet of ings: a semantic-enhanced approach,” Semantic
Web, vol. 10, no. 1, pp. 183–204, 2019.
[7] M. Mafi, W. Izquierdo, M. Cabrerizo et al., “Survey on mixed
impulse and Gaussian denoising filters,” IET Image Process-
ing, vol. 14, no. 16, pp. 4027–4038, 2020.
[8] C. Wang, “IoT anomaly detection method in intelligent
manufacturing industry based on trusted evaluation,” e
International Journal of Advanced Manufacturing Technology,
vol. 107, no. 3, pp. 993–1005, 2020.
[9] M. Yacin Sikkandar, B. A. Alrasheadi, N. B. Prakash,
G. R. Hemalakshmi, A. Mohanarathinam, and K. Shankar,
“Deep learning based an automated skin lesion segmentation
and intelligent classification model,” Journal of Ambient In-
telligence and Humanized Computing, vol. 12, no. 3,
pp. 3245–3255, 2021.
[10] D. Li, Z. Cai, L. Deng, and X. Yao, “IoT complex commu-
nication architecture for smart cities based on soft computing
models,” Soft Computing, vol. 23, no. 8, pp. 2799–2812, 2019.
[11] H. L. Nguyen, D. T. Vu, and J. J. Jung, “Knowledge graph
fusion for smart systems: a Survey,” Information Fusion,
vol. 61, pp. 56–70, 2020.
[12] S. Sankaranarayanan, M. Prabhakar, S. Satish, P. Jain,
A. Ramprasad, and A. Krishnan, “Flood prediction based on
weather parameters using deep learning,” Journal of Water
and Climate Change, vol. 11, no. 4, pp. 1766–1783, 2020.
[13] Y. Sun, S. He, and F. Tong, “Media access control for nar-
rowband internet of things: a survey,” Encyclopedia of
Wireless Networks, pp. 795–799, 2020.
[14] P. Wang, L. T. Yang, J. Li, J. Chen, and S. Hu, “Data fusion in
cyber-physical-social systems: state-of-the-art and perspec-
tives,” Information Fusion, vol. 51, pp. 42–57, 2019.
[15] B. P. L. Lau, S. H. Marakkalage, Y. Zhou et al., “A survey of
data fusion in smart city applications,” Information Fusion,
vol. 52, pp. 357–374, 2019.
[16] S. Vidyashree, “Smart shopping using android application,”
Journal of Research Proceedings, vol. 1, no. 2, pp. 243–252,
[17] R. F. Mansour, J. Escorcia-Gutierrez, M. Gamarra, D. Gupta,
O. Castillo, and S. Kumar, “Unsupervised deep learning based
variational autoencoder model for COVID-19 diagnosis and
classification,” Pattern Recognition Letters, vol. 151,
pp. 267–274, 2021.
[18] Y. Han, C. J. Zhang, and L. Wang, “Industrial IoT for in-
telligent steelmaking with converter mouth flame spectrum
information processed by deep learning,” IEEE Transactions
on Industrial Informatics, vol. 16, no. 4, pp. 2640–2650, 2019.
[19] S. H. Haji and A. B. Sallow, “IoT for smart environment
monitoring based on Python: a review,” Asian Journal of
Research in Computer Science, pp. 57–70, 2021.
[20] R. Chandra Shit, “Crowd intelligence for sustainable futuristic
intelligent transportation system: a review,” Iet Intelligent
Transport Systems, vol. 14, no. 6, pp. 480–494, 2020.
[21] G. A. Fotso Kamga, L. Bitjoka, T. Akram, A. Mengue Mbom,
S. Rameez Naqvi, and Y. Bouroubi, “Advancements in satellite
image classification: methodologies, techniques, approaches
and applications,” International Journal of Remote Sensing,
vol. 42, no. 20, pp. 7662–7722, 2021.
Computational Intelligence and Neuroscience 11
... e globalization of knowledge and the rising tide of scientific and technological innovation have made innovation capabilities the main driver of competitive advantage for technological agents. e creation and flow of knowledge are a key factor for national economic growth at the macro level while determining the competitiveness of scientific research at the micro level [1]. Against the background of accelerated technological change and knowledge flow, global knowledge subjects gradually present a state of open development of interconnection, interdependence, and competition among themselves [2]. ...
Full-text available
This paper provides an in-depth analysis and study of the decomposition of sports biology of athletics events through multimedia image acquisition techniques. The proposed design uses SOPC hardware-software collaboration technology, which makes full use of the parallelism of hardware as well as the flexibility of software to essentially improve the speed of image processing and greatly improve the efficiency of image processing. The high speed and parallelism of hardware are used to realize the image acquisition part with high timing requirements and the preprocessing algorithm with a large number of operations and strong repetition, in addition to the difficulties of many initialization configuration parameters of each peripheral. Additionally, the need for frequent adjustment of peripheral working parameters is solved by making full use of the scalability and flexibility of software, and the control signals output by software can coordinate the work of each hardware module to ensure that each module of the system cooperates. It is also possible to coordinate the work of each hardware module through the control signal output from the software to ensure that each module of the system cooperates and operates in an orderly and efficient manner and provides the possibility of realizing a higher level of the image processing algorithm. The system includes four main factors: swinging action, supporting action, vacating action, and speed rhythm. Based on this system, 14 key kinematic indicators were selected to reflect the technical status of the youth walkers. The landing angle decreased with the increase of speed, and the landing technique was insufficient; it had better control of the large and small arm angle, and their stride width was open, while the other four athletes showed increased tension in the hip joint and upper limb with the increase of walking speed, as well as the wrong action of forward and internal rotation of the large and small arms.
Full-text available
Air pollution, water pollution, and radiation pollution are significant environmental factors that need to be addressed. Proper monitoring is crucial with the goal that by preserving a healthy society, the planet can achieve sustainable development. With advancements in the internet of things (IoT) and the improvement of modern sensors, environmental monitoring has evolved into a smart environment monitoring (SEM) system in recent years. This article aims to have a critical overview of significant contributions and SEM research, which include monitoring the quality of air , water pollution, radiation pollution, and agricultural systems. The review is divided based on the objectives of applying SEM methods, analyzing each objective about the sensors used, machine learning, and classification methods. Moreover, the authors have thoroughly examined how advancements in sensor technology, the Internet of Things, and machine learning methods have made environmental monitoring into a truly smart monitoring system.
Full-text available
This study presents a comprehensive survey on mixed impulse and Gaussian denoising filters which are applied to an image in order to gauge the effects of this type of noise combination and to then determine optimal ways that can overcome such effects. The random noise model considered in this survey is the combined effect of impulse (salt and pepper) and Gaussian noise. After describing the noise models, the denoising filters which are applied to the images are classified and explained according to their design structure, the type of filters they use, the noise level they could overcome, and the limitations they face. This survey covers all related denoising methods and provides an assessment of the strengths and practical limitations of the different classes of denoising filters.
Full-text available
Based on the n-ary counting system, combined with the matrix semi-tensor product theory and Hilbert curve, a chaotic image encryption algorithm is designed. Different from the traditional encryption method, the algorithm proposed in this paper is an encryption algorithm with scrambling and diffusion at the same time. First, the pixel value is converted from decimal to n-ary. In the n-ary counting system, the plaintext image is randomly divided into some groups, and the Hilbert curve is used for scrambling to each group. The blocks are converted into scrambled images, so that the scrambling and diffusion can be carried out at the same time. Then, in order to improve the security of the algorithm, another round of diffusion is carried out based on matrix semi-tensor product mechanism. Chaotic sequence is generated by Chen system. This chaotic sequence performs matrix semi-tensor product operation with the first round of encrypted image, and generate second encrypted images. Finally, this encryption method is applied to color image encryption. Compared with some representative algorithms, the experimental results show that the algorithm proposed in this paper is secure and it can resist common attacks.
Full-text available
Internet of Medical Things (IoMT) includes interconnected sensors, wearable devices, medical devices, and clinical systems. At the same time, skin cancer is a commonly available type of cancer that exists all over the globe. This study projects a new segmentation based classification model for skin lesion diagnosis by combining a GrabCut algorithm and Adaptive Neuro-Fuzzy classifier (ANFC) model. The proposed method involves four main steps: preprocessing, segmentation, feature extraction, and classification. Initially, the preprocessing step is carried out using a Top hat filter and inpainting technique. Then, the Grabcut algorithm is used to segment the preprocessed images. Next, the feature extraction process takes place by the use of a deep learning based Inception model. Finally, an adaptive neuro-fuzzy classifier (ANFC) system gets executed to classify the dermoscopic images into different classes. The proposed model is simulated using a benchmark International Skin Imaging Collaboration (ISIC) dataset and the results are examined interms of accuracy, sensitivity and specificity. The proposed model exhibits better identification and classification of skin cancer. For examining the effective outcome of the projected technique, an extensive comparison of the presented method with earlier models takes place. The experimental values indicated that the proposed method has offered a maximum sensitivity of 93.40%, specificity of 98.70% and accuracy of 97.91%.
Full-text available
The rapid change of climate, population explosion, and reduction of arable lands are calling for new approaches to ensure sustainable agriculture and food supply for the future. Greenhouse agriculture is considered to be a viable alternative and sustainable solution, which can combat the future food crisis by controlling the local environment and growing crops all year round even in harsh outdoor conditions. However, greenhouse farms persist many challenges for efficient operation and management. The evolving Internet of Things (IoT) technologies, which encompass the smart sensors, devices, network topologies, big data analytics, and intelligent decision, is believed to be the solution in addressing the key challenges facing the greenhouse farming, such as greenhouse local climate control, crop growth monitoring, crop harvesting and etc. This paper reviews the current greenhouse cultivation technologies as well as the stateof-the-art of IoT technologies for smart greenhouse farms. The paper also highlights the major challenges that need to be addressed.
At present times, COVID-19 has become a global illness and infected people has increased exponentially and it is difficult to control due to the non-availability of large quantity of testing kits. Artificial intelligence (AI) techniques including machine learning (ML), deep learning (DL), and computer vision (CV) approaches find useful for the recognition, analysis, and prediction of COVID-19. Several ML and DL techniques are trained to resolve the supervised learning issue. At the same time, the potential measure of the unsupervised learning technique is quite high. Therefore, unsupervised learning techniques can be designed in the existing DL models for proficient COVID-19 prediction. In this view, this paper introduces a novel unsupervised DL based variational autoencoder (UDL-VAE) model for COVID-19 detection and classification. The UDL-VAE model involved adaptive Wiener filtering (AWF) based preprocessing technique to enhance the image quality. Besides, Inception v4 with Adagrad technique is employed as a feature extractor and unsupervised VAE model is applied for the classification process. In order to verify the superior diagnostic performance of the UDL-VAE model, a set of experimentation was carried out to highlight the effective outcome of the UDL-VAE model. The obtained experimental values showcased the effectual results of the UDL-VAE model with the higher accuracy of 0.987 and 0.992 on the binary and multiple classes respectively.
Segmentation and classification are two imperative, yet challenging tasks in image analysis for remote-sensing applications. In the former, an image is divided into spatially continuous, disjoint, and homogeneous regions, called clusters, in terms of their various properties: shape, intensity, texture, colour, contrast, etc. Classification, on the other hand, is applied later in the process, to recognize or categorize individual objects or targets. Each task plays an important role in the refinement and enhancement of the various utilizations of remote sensing images. Driven by recent progress in earth observation sensor technology, satellite image classification systems for earth-observation applications have seen significant growth and progress. This growth has led to a notable increase in the number of published materials in these areas. We present an overview of the horizons that the modern remote sensing domain promises in terms of the efficient classification processing of satellite imagery. We begin by defining remote sensing, specifically in the context of its potential application areas, and highlight the importance of pre-processing and feature extraction steps’ in accurate classification. Various works have proposed novel segmentation, feature extraction/selection, and classification methods; these have been collected and duly reported in this work. The deep learning classification method has been given special attention due to its relatively limited dependence on training data, its wide spectrum of applications, and its ability to autonomously classify images with higher accuracy. We conclude by presenting a critical evaluation of the important contributions in this domain.
In the Internet of Things (IoT) era, billions of sensors and devices collect and process data from the environment, transmit them to cloud centers, and receive feedback via the internet for connectivity and perception. However, transmitting massive amounts of heterogeneous data, perceiving complex environments from these data, and then making smart decisions in a timely manner are difficult. Artificial intelligence (AI), especially deep learning, is now a proven success in various areas including computer vision, speech recognition, and natural language processing. AI introduced into the IoT heralds the era of artificial intelligence of things (AIoT). This paper presents a comprehensive survey on AIoT to show how AI can empower the IoT to make it faster, smarter, greener, and safer. Specifically, we briefly present the AIoT architecture in the context of cloud computing, fog computing, and edge computing. Then, we present progress in AI research for IoT from four perspectives: perceiving, learning, reasoning, and behaving. Next, we summarize some promising applications of AIoT that are likely to profoundly reshape our world. Finally, we highlight the challenges facing AIoT and some potential research opportunities.
Internet of things (IoT) is a promising solution to connect and access every device through internet. Every day the device count increases with large diversity in shape, size, usage and complexity. Since IoT drive the world and changes people lives with its wide range of services and applications. However, IoT provides numerous services through applications, it faces severe security issues and vulnerable to attacks such as sinkhole attack, eaves dropping, denial of service attacks, etc., Intrusion detection system is used to detect such attacks when the network security is breached. This research work proposed an intrusion detection system for IoT network and detect different types of attacks based on hybrid convolutional neural network model. Proposed model is suitable for wide range of IoT applications. Proposed research work is validated and compared with conventional machine learning and deep learning model. Experimental result demonstrate that proposed hybrid model is more sensitive to attacks in the IoT network.
The emergence of various disruptive technologies such as big data, Internet of Things, and artificial intelligence have instigated our society to generate enormous volumes of data. The effective, efficient, and transparent capture and fusion of knowledge from a massive amount data is becoming an increasingly popular and crucial topic. In this study, we aim to provide a broad, complete, and systematic overview of the definitions and challenges of the knowledge graph fusion, which represents a holistic approach for integrating, enhancing, and unifying knowledge graphs. Further, advanced techniques for handling knowledge graph fusion along with the pragmatic smart systems leveraging it are discussed as a part of multiple perspectives. We believe that this survey study can be used as a potential reference for system practitioners and researchers in surpassing current obstacles as well as shaping their future direction.