Content uploaded by Weiming Dong
Author content
All content in this area was uploaded by Weiming Dong on Jan 23, 2014
Content may be subject to copyright.
A preview of the PDF is not available
Image zooming using directional cubic convolution interpolation
IET Image Processing
Abstract
Image-zooming is a technique of producing a high-resolution image from its low-resolution counterpart. It is also called image interpolation because it is usually implemented by interpolation. Keys' cubic convolution (CC) interpolation method has become a standard in the image interpolation field, but CC interpolates indiscriminately the missing pixels in the horizontal or vertical direction and typically incurs blurring, blocking, ringing or other artefacts. In this study, the authors propose a novel edge-directed CC interpolation scheme which can adapt to the varying edge structures of images. The authors also give an estimation method of the strong edge for a missing pixel location, which guides the interpolation for the missing pixel. The authors' method can preserve the sharp edges and details of images with notable suppression of the artefacts that usually occur with CC interpolation. The experiment results demonstrate that the authors'method outperforms significantly CC interpolation in terms of both subjective and objective measures.
... Cropping is often used to remove unnecessary background elements, enhance composition, adjust perspective, or focus on specific areas or subjects within the image. This process can lead to variations in image size due to the removal of different sections [15,16]. ...
This paper introduces a novel approach to pavement material crack detection, classification, and segmentation using advanced deep learning techniques, including multi-scale feature aggregation and transformer-based attention mechanisms. The proposed methodology significantly enhances the model’s ability to handle varying crack sizes, shapes, and complex pavement textures. Trained on a dataset of 10,000 images, the model achieved substantial performance improvements across all tasks after integrating transformer-based attention. Detection precision increased from 88.7% to 94.3%, and IoU improved from 78.8% to 93.2%. In classification, precision rose from 88.3% to 94.8%, and recall improved from 86.8% to 94.2%. For segmentation, the Dice Coefficient increased from 80.3% to 94.7%, and IoU for segmentation advanced from 74.2% to 92.3%. These results underscore the model’s robustness and accuracy in identifying pavement cracks in challenging real-world scenarios. This framework not only advances automated pavement maintenance but also provides a foundation for future research focused on optimizing real-time processing and extending the model’s applicability to more diverse pavement conditions.
... This is because it does not take into account the influence of other neighboring pixels on the target interpolated point, especially when applied to higher-resolution images. To address the issue of aliasing artifacts caused by the neglect of interactions between adjacent pixels in nearest-neighbor interpolation, researchers introduced bilinear interpolation [160][161][162]. ...
Carbon dioxide is one of the most influential greenhouse gases affecting human life. CO2 data can be obtained through three methods: ground-based, airborne, and satellite-based observations. However, ground-based monitoring is typically composed of sparsely distributed stations, while airborne monitoring has limited coverage and spatial resolution; they cannot fully reflect the spatiotemporal distribution of CO2. Satellite remote sensing plays a crucial role in monitoring the global distribution of atmospheric CO2, offering high observation accuracy and wide coverage. However, satellite remote sensing still faces spatiotemporal constraints, such as interference from clouds (or aerosols) and limitations from satellite orbits, which can lead to significant data loss. Therefore, the reconstruction of satellite-based CO2 data becomes particularly important. This article summarizes methods for the reconstruction of satellite-based CO2 data, including interpolation, data fusion, and super-resolution reconstruction techniques, and their advantages and disadvantages, it also provides a comprehensive overview of the classification and applications of super-resolution reconstruction techniques. Finally, the article offers future perspectives, suggesting that ideas like image super-resolution reconstruction represent the future trend in the field of satellite-based CO2 data reconstruction.
The single image super-resolution (SISR) technique turns a low resolution (LR) image into a clear image. Many techniques, including edge-directed, dictionaries-based, and polynomial-based are offered for SISR. However certain blurring effects are formed when the LR image is converted to a higher resolution (HR) image. Recently, deep-learning methods based on convolutional neural networks (CNN) have emerged and are being used for super resolution. It is capable of restoring comprehensive information in HR images. However, these techniques are computationally costly and have significant memory needs, making them difficult to adapt to real-time settings. To overcome the aforementioned issue, iterative back projection-based dual path CNN (IBPDPNet) is utilized, with the LR image shallow and deep features extracted independently using the improved residual block and the recursive dilated convolution block. To reduce the blurring effects after upscaling, one lost detail information of the LR image in terms of back projecting error is added to the deep, and shallow features. To produce the essential features from the retrieved features channel-wise attention is applied followed by pixel shuffle for feature up-sampling. By implementing the aforesaid technique, the HR image retained with edge, texture, and geometric regularity is generated. The efficiency of the proposed strategy is better in terms of subjective and objective quality compared to the existing approach.
In order to obtain super-resolution images with richer details and clearer textures, a method for image super-resolution reconstruction using dense residual attention and multi-scale fusion is proposed. First, different scale convolutions are used to fully extract shallow features of the image; then high-frequency features of the image are extracted through one three-layer cascaded multi-scale feature fusion and dense residual attention module, and the reuse of feature map is achieved; finally, residual branches are used to introduce shallow features and high-frequency features of each channel image, and the high-resolution images are reconstructed through up-sampling and sub-pixel convolution. The test results on the Set5, Set14, Bsd100, and Urban100 datasets show that the PSNR and SSIM of our model are superior to most current algorithms, especially in the case of [Formula: see text] reconstruction results. PSNR has improved by 0.2 dB on the Set5 and Bsd100 datasets, and the algorithm has a better subjective visual effect.
In this work, we introduce a novel image zooming methodology that transitions from a nonadaptive Sin‐based approach to an adaptive Sinc‐based zooming technique. The two techniques base their theoretical foundation on the Whittaker–Shannon interpolation formula and the Nyquist theorem. The evolution into adaptive Sinc‐based zoom is accomplished through the use of two novel concepts: (1) the pixel‐local scaled k‐space and (2) the k‐space filtering sigmoidal function. The pixel‐local scaled k‐space is the standardized and scaled k‐space magnitude of the image to zoom. The k‐space filtering sigmoidal function scales the pixel‐local scaled k‐space values into the numerical interval [0, 1]. Using these two novel concepts, the Whittaker–Shannon interpolation formula is elaborated and used to zoom images. Zooming is determined by the shape of the Sinc functions in the Whittaker–Shannon interpolation formula, which, in turn, depends on the combined effect of the pixel‐local scaled k‐space, the sampling rate, and the k‐space filtering sigmoidal function. The primary outcome of this research demonstrates that the Whittaker–Shannon interpolation formula can achieve successful zooms for values of the sampling rate significantly greater than the bandwidth. Conversely, when the sampling rate is much greater than the bandwidth, the nonadaptive technique fails to perform the zoom correctly. The conclusion is that the k‐space filtering sigmoidal function is identified as the crucial parameter in the adaptive Sinc‐based zoom technique. The implications of this research extend to Sinc‐based image zooming applications.
The development of infrared image detection technology has improved the real-time performance and safety of fault diagnosis. To address the problem of insufficient features caused by the low resolution of infrared images, a fault diagnosis model for induction motor infrared images based on Dual-Stream Attention Convolutional (DSAC) is proposed. Firstly, this model employs a dual-stream convolutional neural network to extract spatial features in the X and Y directions of the infrared images separately. This dual-stream structure allows the network to simultaneously learn and utilize information from both X and Y directions, enabling a more comprehensive capture of temperature distribution and variation trends in the infrared images. Then, a convolutional attention mechanism is introduced to assign weights to the obtained features. The attention maps generated by the convolutional attention layer enhance key features while suppressing unimportant information, thereby enhancing the focusing performance of the DSAC model on critical features. Finally, the weighted dual-stream features are fused to achieve the goal of fault diagnosis for induction motor infrared images. The DSAC model is validated using the induction motor infrared image dataset from Babol Noshirvani University of Technology, demonstrating excellent diagnostic accuracy and speed on small sample datasets, providing a feasible solution for fault diagnosis of induction motor infrared images.
This article reports a novel adaptive Sinc-based zooming technique. The theoretical basis of the zooming technique is the Whittaker–Shannon interpolation formula and the Nyquist–Shannon sampling theorem. The technique can be used to zoom out, zoom in, and reconstruct two-dimensional images with no zoom. The technique introduces a novel concept called pixel-local-scaled k-space, which is calculated as follows. The k-space magnitude and the bandwidth of the image to zoom are calculated. The bandwidth of the frequency components of the image to zoom is estimated by the difference between the maximum and the minimum numerical value of the k-space magnitude. The ratio between the standardised k-space magnitude of the image and the bandwidth is calculated pixel-by-pixel, and is called pixel-local-scaled k-space. The signal is reconstructed using a finite convolution between image pixel intensities and Sinc functions. The zooming factor is consequential to the numerical difference between the pixel-local-scaled k-space and the sampling rate. The novelty of this zooming technique is to be adaptive because the pixel-local-scaled k-space is a pixel-by-pixel map, that is, an image with the same number of pixels as the departing image. Adaptive Sinc-based zooming is validated through the comparison with non-adaptive Sinc-based zooming, and its smoothing effect is compared to the smoothing of a variety of image space and z-space filters. The adaptive Sinc-based zooming technique achieves image zooming reconstruction using sampling rates within an extended range versus the range allowed by non-adaptive Sinc-based zooming.
Finding the optimal implementation of calculations is one of the most critical challenges in image processing programming. Halide is a domain-specific language for high-performance image processing. Its auto-scheduler is a helpful tool for solving this problem; however, its scheduling is not a panacea for complex flows. In this paper, we evaluate the performance of the auto-scheduler by comparing it to hand-manually implemented C++ codes. The algorithm used for comparison is Directional cubic convolution interpolation (DCCI), whose computation schedule is challenging to optimize. We evaluate three auto-schedulers: Adams et al.'s, Li et al.'s, and Mulla-pudi et al.s. Experimental results show that the performance of the schedule generated by Adams' method is comparable to that of the hand-implemented C++ code.
We propose a new wavelet domain image interpolation scheme based
on statistical signal estimation. A linear composite MMSE estimator is
constructed to synthesize the detailed wavelet coefficients as well as
to minimize the mean squared error for high-resolution signal recovery.
Based on a discrete time edge model, we use low-resolution information
to characterize local intensity changes and perform resolution
enhancement accordingly. A linear MMSE estimator follows to minimize the
estimation error. Local image statistics are involved in determining the
spatially adaptive optimal estimator. With knowledge of edge behavior
and local signal statistics, the composite estimation is able to enhance
important edges and to maintain the intensity consistency along edges.
Strong improvement in both the visual quality and the PSNRs of the
interpolated images has been achieved by the proposed estimation scheme
Efficient algorithms for the continuous representation of a discrete signal in terms of B-splines (direct B-spline transform) and for interpolative signal reconstruction (indirect B-spline transform) with an expansion factor m are described. Expressions for the z-transforms of the sampled B-spline functions are determined and a convolution property of these kernels is established. It is shown that both the direct and indirect spline transforms involve linear operators that are space invariant and are implemented efficiently by linear filtering. Fast computational algorithms based on the recursive implementations of these filters are proposed. A B-spline interpolator can also be characterized in terms of its transfer function and its global impulse response (cardinal spline of order n). The case of the cubic spline is treated in greater detail. The present approach is compared with previous methods that are reexamined from a critical point of view. It is concluded that B-spline interpolation correctly applied does not result in a loss of image resolution and that this type of interpolation can be performed in a very efficient manner
We describe a spatially adaptive algorithm for image interpolation. The algorithm uses a wavelet transform to extract information about sharp variations in the low-resolution image and then implicitly applies interpolation which adapts to the image local smoothness/singularity characteristics. The proposed algorithm yields images that are sharper compared to several other methods that we have considered in this paper. Better performance comes at the expense of higher complexity.
This chapter presents a survey of interpolation and resampling techniques in the context of exact, separable interpolation of regularly sampled data. In this context, the traditional view of interpolation is to represent an arbitrary continuous function as a discrete sum of weighted and shifted synthesis functions—in other words, a mixed convolution equation. An important issue is the choice of adequate synthesis functions that satisfy interpolation properties. Examples of finite-support ones are the square pulse (nearest-neighbor interpolation), the hat function (linear interpolation), the cubic Keys' function, and various truncated or windowed versions of the sinc function. On the other hand, splines provide examples of infinite-support interpolation functions that can be realized exactly at a finite, surprisingly small computational cost. We discuss implementation issues and illustrate the performance of each synthesis function. We also highlight several artifacts that may arise when performing interpolation, such as ringing, aliasing, blocking and blurring. We explain why the approximation order inherent in the synthesis function is important to limit these interpolation artifacts, which motivates the use of splines as a tunable way to keep them in check without any significant cost penalty.
We present a simple, original method to improve piecewise-linear interpolation with uniform knots: we shift the sampling knots by a fixed amount, while enforcing the interpolation property. We determine the theoretical optimal shift that maximizes the quality of our shifted linear interpolation. Surprisingly enough, this optimal value is nonzero and close to 1/5. We confirm our theoretical findings by performing several experiments: a cumulative rotation experiment and a zoom experiment. Both show a significant increase of the quality of the shifted method with respect to the standard one. We also observe that, in these results, we get a quality that is similar to that of the computationally more costly "high-quality" cubic convolution.
Preserving edge structures is a challenge to image interpolation algorithms that reconstruct a high-resolution image from a low-resolution counterpart. We propose a new edge-guided nonlinear interpolation technique through directional filtering and data fusion. For a pixel to be interpolated, two observation sets are defined in two orthogonal directions, and each set produces an estimate of the pixel value. These directional estimates, modeled as different noisy measurements of the missing pixel are fused by the linear minimum mean square-error estimation (LMMSE) technique into a more robust estimate, using the statistics of the two observation sets. We also present a simplified version of the LMMSE-based interpolation algorithm to reduce computational cost without sacrificing much the interpolation performance. Experiments show that the new interpolation techniques can preserve edge sharpness and reduce ringing artifacts.
Assumptions about image continuity lead to oversmoothed edges in common image interpolation algorithms. A wavelet-based interpolation method that imposes no continuity constraints is introduced. The algorithm estimates the regularity of edges by measuring the decay of wavelet transform coefficients across scales and preserves the underlying regularity by extrapolating a new subband to be used in image resynthesis. The algorithm produces visibly sharper edges than traditional techniques and exhibits an average peak signal-to-noise ratio (PSNR) improvement of 2.5 dB over bilinear and bicubic techniques.
In this paper, we present a nonlinear interpolation scheme for still image resolution enhancement. The algorithm is based on a source model emphasizing the visual integrity of detected edges and incorporates a novel edge fitting operator that has been developed for this application. A small neighborhood about each pixel in the low-resolution image is first mapped to a best-fit continuous space step edge. The bilevel approximation serves as a local template on which the higher resolution sampling grid can then be superimposed (where disputed values in regions of local window overlap are averaged to smooth errors). The result is an image of increased resolution with noticeably sharper edges and, in all tried cases, lower mean-squared reconstruction error than that produced by linear techniques.
This paper presents an edge-directed image interpolation algorithm. In the proposed algorithm, the edge directions are implicitly estimated with a statistical-based approach. In opposite to explicit edge directions, the local edge directions are indicated by length-16 weighting vectors. Implicitly, the weighting vectors are used to formulate geometric regularity (GR) constraint (smoothness along edges and sharpness across edges) and the GR constraint is imposed on the interpolated image through the Markov random field (MRF) model. Furthermore, under the maximum a posteriori-MRF framework, the desired interpolated image corresponds to the minimal energy state of a 2-D random field given the low-resolution image. Simulated annealing methods are used to search for the minimal energy state from the state space. To lower the computational complexity of MRF, a single-pass implementation is designed, which performs nearly as well as the iterative optimization. Simulation results show that the proposed MRF model-based edge-directed interpolation method produces edges with strong geometric regularity. Compared to traditional methods and other edge-directed interpolation methods, the proposed method improves the subjective quality of the interpolated edges while maintaining a high PSNR level.
In the problem of image interpolation, most of the difficulties arise in areas around edges and sharp changes. Around edges, many interpolation methods tend to smooth and blur image detail. Fortunately, most of the signal information is often carried around edges and areas of sharp changes and can be used to predict these missing details from a sampled image. A method for adding image detail based on the cone of influence, the evolution of the wavelet coefficients across scales, is presented.