R. Devi Priya’s research while affiliated with Coimbatore Institute of Engineering and Technology and other places

The recent and fast improvements in communication networks and digital technologies have facilitated the easy transmission and storage of multimedia information on various network platforms. However, in today’s research, issues relating to the safety concerned with the large amount of information have emerged as a crucial topic due to the internet openings. Since images are regarded as significant components of network communication, the information security of images is highly vital. In this paper, an image processing concept along with authenticated cloud storage is proposed which includes the enhanced processing of input images and their secured storage. The processed input images undergo authentication procedures in which the algorithm used to encrypt and decrypt the image information is the Nth Degree Truncated Polynomial Ring Unit (NTRU). It performs a rapid generation of 128 bit private key and verification achieving improved performance and thereby preventing the unauthorized access of information. The processed image is encrypted and stored in the cloud using the proposed algorithm. Only an authorized person is permitted to do accessible secret measures to decrypt the document from the cloud. Meanwhile, the proposed schemes show more efficient and secure one to analyze the security, which are done in the assessment of performance of the system generating improved values of peak signal-to-noise ratio (PSNR) and mean-square error (MSE) indicating 5.24 and 82.10, respectively. The proposed work exhibits an average encryption time of 12.075 s and decryption time of 16.2 s, respectively.

The most common form of communication is speech. Speech Emotion Recognition is a demanding study area that aims to identify human emotions from speech information (SER). Capturing the emotion from only speech is a difficult challenge. The proper selection of features in relation with both the time and frequency domains together is necessary to produce optimized results. In this paper, four emotional states: Happy, Sad, Anger and Neutral from speech are recognized by using two classifiers. The speech utterances are taken from the Poland Corpus (Database of Polish Emotional Speech). The explored features include Energy, Pitch, Zero-crossing rate and Mel-Frequency Cepstrum Coefficients (MFCC). Performance is compared by employing two different classification algorithms namely, Support vector machine (SVMs) and Linear Discriminant Analysis (LDA). The experimental results reveal that SVM offers15% relative accuracy improvement compared to LDA.KeywordsSpeech SignalEmotion RecognitionRecognition RateSupport Vector MachineLinear Discriminant Analysis

Tamil character recognition is an important field of research in pattern recognition and it is a technical challenge than other languages due to similarity and complexity of characters. Stone inscriptions reveal the details of lavishness, lifestyle, economic conditions, culture, and also of the managerial regulations followed by various rulers and dynasties particular to those regions. However, due to the long history of ancient stone inscription, natural erosion and lack of early protection measures, there are lot of noise in the existing ancient stone inscriptions, which create adverse effects on reading these stone inscriptions and their aesthetic appreciation. The research challenge in recognizing Tamil characters is mainly because of the characters with a number of holes, loops and curves. The number of letters in Tamil language is higher when compared to other languages. Even though there are various approaches provided by the researchers, challenges and issues still prevail in recognition of tamil text in stone inscriptions. In the existing systems, detection algorithms fail to produce desired accuracy and hence stone inscription recognition using transfer learning, a promising method is proposed in this paper. Lion Optimization Algorithm (LOA) is applied to optimize brightness and contrast and then stone inscription images are pre-processed for noise removal and then each character is separated by identifying contours. Characters are recognized using Transfer Learning (TL), a Deep Convolution Neural Network-based multi classification approach. The proposed hybrid model Self-Adaptive Lion Optimization Algorithm with Transfer Learning (SLOA-TL) when implemented in images of stone inscriptions achieves better accuracy and speed than other existing methods. It serves as an efficient design for recognition of tamil characters in stone inscriptions and preserving tamil traditional knowledge.

Today’s datasets are usually very large with many features and making analysis on such datasets is really a tedious task. Especially when performing classification, selecting attributes that are salient for the process is a brainstorming task. It is more difficult when there are many class labels for the target class attribute and hence many researchers have introduced methods to select features for performing classification on multi-class attributes. The process becomes more tedious when the attribute values are imbalanced for which researchers have contributed many methods. But, there is no sufficient research to handle extreme imbalance and feature selection together and hence this paper aims to bridge this gap. Here Particle Swarm Optimization (PSO), an efficient evolutionary algorithm is used to handle imbalanced dataset and feature selection process is also enhanced with the required functionalities. First, Multi-objective Particle Swarm Optimization is used to transform the imbalanced datasets into balanced one and then another version of Multi-objective Particle Swarm Optimization is used to select the significant features. The proposed methodology is applied on eight multi-class extremely imbalanced datasets and the experimental results are found to be better than other existing methods in terms of classification accuracy, G mean, F measure. The results validated by using Friedman test also confirm that the proposed methodology effectively balances the dataset with less number of features than other methods.

Classification algorithms and their preprocessing operations usually performs on feature selection on homogeneous or heterogeneous attributes, binary or multi-class labels separately. Only very few methods attempt to perform feature selection on datasets with heterogeneous multi-class attributes. In order to bridge this gap with better classification performance, the paper proposes a Tri-staged Feature Selection (TFS) methodology which performs (i) Feature selection using Kruskal Wallis test (ii) Refinement of feature selection using a new Memetic Algorithm with local beam search and genetic algorithm operations and (iii) Further refinement of feature selection using Cuckoo Search algorithm. Proper tradeoff between both exploration and exploitation is maintained in the proposed method. The experimental results on 12 datasets show that the proposed method is better than that of state-of-the-art methods used for feature selection in terms of multi-class accuracy, hamming loss, ranking loss, normalized coverage and convergence rate for multi-class heterogeneous datasets.

COVID-19 caused by novel Corona virus emerges as a pandemic threat to the entire humanity. People who suffer from malnutrition and lower immunity to fight against the virus are affected severely. From the corona statistics, it is reported children below 13 years, elder people and pregnant women are easily affected by Corona virus than other age group and their mortality rate is also higher. With no vaccine available till now, the only treatment option for the entire world is to have nutritious food and maintain better immunity against Corona virus. This issue becomes more severe in tribal regions of Tamilnadu where they have very low access to medical facilities. To create a healthy generation capable of surviving such infectious illness and other illness, balanced healthy nutrition without under-nourishment is very much essential. Developing such healthy individual starts during the time of pregnancy and hence, pregnant women need to be well-nourished, since it directly affects health of the infants. Malnutrition is a common problem to people of all ages. But, focus of the paper is only on pregnant women, married women who are yet to get pregnant and children under five years of age. Nutritional status of other people are not included in this study. In all the five districts, majority of tribal people live in forests with less resources, financial conditions, educational knowledge and exposure to modern methods available for pregnancy and infant development.KeywordsMalnutritionMachine learning techniquesClassificationFeature selection

In Indian scenario, agriculture is the main source of economy for most of the people. Cotton serves as one of the primary crops that is cultivated in India and provide income to many farmers in the village. Yet, its production is adversely affected by pests, insects, insufficient nutrition and weather conditions. Leaves get heavily infected and detection of diseases that commonly affect leaves should be identified at the early stages to prevent reduction in yield. Appropriate measures like spraying of insecticides, pesticides can be taken thereafter and the crops can be saved. Some machine learning methods are already available to detect the diseases in cotton leaves. Yet, they do not show better accuracy as expected in detection and classification of diseases. This paper proposes an algorithm called Augmented Faster R-CNN (AFR-CNN) by combining Faster R-CNN, an efficient deep learning algorithm with suitable data augmentation techniques like rotation, blur transformation, flipping and GAN. The proposed algorithm is implemented with 4000 images of cotton leaves with diseases. From the implementation results, it is understood that proposed AFR-CNN superior to existing methods for cotton leaf disease detection. The significance of preprocessing and augmentation methods is emphasized with experimental results. Appropriate remedies are also suggested so that the farmers can take strategic decisions.KeywordsDetection of cotton leaves detectionCycle-GANAugmentation methodsFaster R-CNN