Edge Pixels Research Articles

Kruskal Szekeres generative adversarial network augmented deep autoencoder for colorectal cancer detection.

Cancer involves abnormal cell growth, with types like intestinal and oesophageal cancer often diagnosed in advanced stages, making them hard to cure. Symptoms are like burning sensations in the stomach and swallowing difficulties are specified as colorectal cancer. Deep learning significantly impacts the medical image processing and diagnosis, offering potential improvements in accuracy and efficiency. The Kruskal Szekeres Generative Adversarial Network Augmented Deep Autoencoder (KSGANA-DA) is introduced for early colorectal cancer detection and it comprises two stages; Initial stage, data augmentation uses Affine Transform via Random Horizontal Rotation and Geometric Transform via Kruskal-Szekeres that coordinates to improve the training dataset diversity, boosting detection performance. The second stage, a Deep Autoencoder Anatomical Landmark-based Image Segmentation preserves edge pixel spatial locations, improving precision and recall for early boundary detection. Experiments validate KSGANA-DA performance and different existing methods are implemented into Python. The results of KSGANA-DA are to provide higher precision by 41%, recall by 7%, and lesser training time by 46% than compared to conventional methods.

High-Quality Damaged Building Instance Segmentation Based on Improved Mask Transfiner Using Post-Earthquake UAS Imagery: A Case Study of the Luding Ms 6.8 Earthquake in China

Unmanned aerial systems (UASs) are increasingly playing a crucial role in earthquake emergency response and disaster assessment due to their ease of operation, mobility, and low cost. However, post-earthquake scenes are complex, with many forms of damaged buildings. UAS imagery has a high spatial resolution, but the resolution is inconsistent between different flight missions. These factors make it challenging for existing methods to accurately identify individual damaged buildings in UAS images from different scenes, resulting in coarse segmentation masks that are insufficient for practical application needs. To address these issues, this paper proposed DB-Transfiner, a building damage instance segmentation method for post-earthquake UAS imagery based on the Mask Transfiner network. This method primarily employed deformable convolution in the backbone network to enhance adaptability to collapsed buildings of arbitrary shapes. Additionally, it used an enhanced bidirectional feature pyramid network (BiFPN) to integrate multi-scale features, improving the representation of targets of various sizes. Furthermore, a lightweight Transformer encoder has been used to process edge pixels, enhancing the efficiency of global feature extraction and the refinement of target edges. We conducted experiments on post-disaster UAS images collected from the 2022 Luding earthquake with a surface wave magnitude (Ms) of 6.8 in the Sichuan Province of China. The results demonstrated that the average precisions (AP) of DB-Transfiner, APbox and APseg, are 56.42% and 54.85%, respectively, outperforming all other comparative methods. Our model improved the original model by 5.00% and 4.07% in APbox and APseg, respectively. Importantly, the APseg of our model was significantly higher than the state-of-the-art instance segmentation model Mask R-CNN, with an increase of 9.07%. In addition, we conducted applicability testing, and the model achieved an average correctness rate of 84.28% for identifying images from different scenes of the same earthquake. We also applied the model to the Yangbi earthquake scene and found that the model maintained good performance, demonstrating a certain level of generalization capability. This method has high accuracy in identifying and assessing damaged buildings after earthquakes and can provide critical data support for disaster loss assessment.

Retinal artery/vein vessel segmentation and measurement for hypertension based on multi-level edge-guided network

AbstractHypertension is a primary risk factor for the onset of cardiocerebrovascular diseases, leading to increased mortality. In the early stages of hypertension, changes in the diameter of the retinal arteries and veins are closely observed. Therefore, the automatic segmentation of these arteries and veins in retinal fundus images is crucial for hypertension monitoring. However, current deep learning based methods still fail to generate semantically consistent segmentation result, especially for narrow blood vessel segments. Moreover, the pixels near the vessel edge are prone to be missclassified by current methods, struggling to obtain edge-preserving segmentation results. These critical issues severely hinder downstream applications, which require accurate measurement of artery/vein diameter. First, to alleviate the issue of lacking semantic consistency for vessel segments, we propose a long-range spatial dependency modeling module to learn to model the long-range spatial dependency. On this basis, a multi-level edge guided spatial aggregation module is further presented to enhance the ability to accurately classify the edge pixels, generating edge-preserving results. Extensive experimental results on the widely used DRIVE dataset and our constructed dataset (HFC-50) show the superiority of the proposed model over state-of-the-art methods. Our method achieves balanced accuracy of 95.7% on DRIVE dataset, outperforming all state-of-art methods. Finally, in order to directly demonstrate the benefit of more accurate retinal artery/vein vessel segmentation for the measurement of the ratio of artery and vein, we quantitatively evaluate the computed artery/vein ratio for hypertension patients, with the artery/vein segmentation results generated by our proposed method.

Saliency and edge features-guided end-to-end network for salient object detection

The rapid development of the Vision Transformer backbones has enabled the capture of feature information with global dependencies, leading to excellent performance in salient object detection tasks. However, it fails to adequately emphasize fine local features of edges, resulting in coarse and blurry edges in the final output. Therefore, this paper proposes a set prediction method for salient object detection. The approach enables the model to simultaneously consider both saliency (salient object) and edge features, achieving an end-to-end edge feature fusion. There is no need to design multiple complex branch structures and multiple training as with other methods. The model integrates random edge neighborhood sampling to enhance the recognition accuracy of local edge features in images. This approach addresses the weak perception of local features by Transformers and the issue encountered during actual training, where edge pixels typically occupy a much smaller proportion of the image compared to the background, resulting in insufficient learning of edge features by the model. The proposed end-to-end model in this paper fuses multiple features, including edge and salient objects. They are extracted within a unified framework while simultaneously outputting edge and salient object maps. Experimental results on six public datasets show that the proposed method significantly improves model performance benchmarks for detecting salient objects.

DAT-Net: Deep Aggregation Transformer Network for automatic nuclear segmentation

Detection and segmentation of cell nuclei in hematoxylin and eosin-stained tissue images pose critical clinical challenges, including intricate backgrounds, variable nuclear appearances, overlapping nuclei, and unclear boundaries. These complexities render automated instance segmentation a difficult research area. Thus, this paper proposes a deep aggregation transformer network for automatic nucleus instance segmentation. The method employs a robust Vision Transformer-based backbone to extract informative features from images. To fuse multi-level information, we propose the group aggregation multi-level feature fusion module, which merges low-level features for shape representation and high-level features for pixel edge segmentation, enhancing overall model performance. The depthwise integrated multilayer perceptron module enhances feature learning by combining the global context-dependent representation of MLP and the local representation of convolution through channel separation. This fosters feature learning across different modes, thereby improving network representation. The bidirectional path interaction feature processing module is introduced for multi-scale information extraction and fusion. This module enhances the entire feature pyramid hierarchy by transferring deep semantic information to shallow layers with robust spatial information via bidirectional paths. The Large kernel Attention module at each layer extensively improves feature extraction through a large receptive field. Our approach outperforms state-of-the-art models, including StarDIST, CPP-Net, TSFD-Net and TransNuSeg, by 7.8%, 6.3%, 1.5% and 3.5% in average panoptic quality for different cell nuclei segmentation categories across tissues of the Pannuke dataset. Furthermore, DAT-Net outperforms all other state-of-the-art models in terms of the F1 score for the five classes of cell nuclei in the PanNuke dataset.

Fuzzy C-Means clustering based selective edge enhancement scheme for improved road crack detection

One of the most serious problems currently in the world is road cracking, which could lower traffic safety and raise the risk of road accidents. A substantial sum of money is spent annually on maintaining and fixing the roads. However, manual crack detection is less accurate and takes longer. The focus of this study is on how artificial intelligence can be applied to image processing to improve crack detection. This study introduces the pixel classification enhancement fuzzy C-Means clustering, a new fuzzy C-Means clustering algorithm. The road crack detection from image is achieved automatically by utilizing second order pixel differences, an exponentially varying intensity dependent edge pixel enhancement scaling factor and intensity-based edge & non-edge fuzzy factors that helps to detect the cracks even in low contrast photos. This method doesn't necessitate training with image datasets. In addition, it can recognize edges or fractures in a range of unseen images. According to standard performance parameters, the experimental results demonstrate that the novel fuzzy C-Means segmentation-based approach performs better than many of the existing methods for contaminated or non-contaminated images in identifying patholes, transverse, and longitudinal cracks from road photos, as well as alligator cracks. The proposed method can assist managers or maintenance engineers in preventing further damage and quickly identifying different types of cracks with less manual labor and expense.

A Microvascular Segmentation Network Based on Pyramidal Attention Mechanism.

The precise segmentation of retinal vasculature is crucial for the early screening of various eye diseases, such as diabetic retinopathy and hypertensive retinopathy. Given the complex and variable overall structure of retinal vessels and their delicate, minute local features, the accurate extraction of fine vessels and edge pixels remains a technical challenge in the current research. To enhance the ability to extract thin vessels, this paper incorporates a pyramid channel attention module into a U-shaped network. This allows for more effective capture of information at different levels and increased attention to vessel-related channels, thereby improving model performance. Simultaneously, to prevent overfitting, this paper optimizes the standard convolutional block in the U-Net with the pre-activated residual discard convolution block, thus improving the model's generalization ability. The model is evaluated on three benchmark retinal datasets: DRIVE, CHASE_DB1, and STARE. Experimental results demonstrate that, compared to the baseline model, the proposed model achieves improvements in sensitivity (Sen) scores of 7.12%, 9.65%, and 5.36% on these three datasets, respectively, proving its strong ability to extract fine vessels.

Cross-Correlation Driven Aerial Image Segmentation: Leveraging Multi-Scale Features and Edge Information

Abstract: Semantic segmentation of remote sensing images is crucial for interpreting these large, rich in information scenes and our study introduces a new method for segmenting remote sensing imagery (RSI) when faced with limited training data and imbalanced classes. Our approach utilizes a unique potential function that merges information from both super pixel segmentation and edge detection. This combination allows the model to effectively analyze features at various scales and reduce the influence of potential errors in super pixel segmentation. Furthermore, the inclusion of edge details extracted via the Sketch token algorithm refines object boundaries, yielding more accurate segmentation results. This work offers a promising solution for achieving reliable interpretation of RSIs in scenarios with limited training data.

RETRACTED: A deep learning approach based on graphs to detect plantation lines

Identifying plantation lines in aerial images of agricultural landscapes is re-quired for many automatic farming processes. Deep learning-based networks are among the most prominent methods to learn such patterns and extract this type of information from diverse imagery conditions. However, even state-of-the-art methods may stumble in complex plantation patterns. Here, we propose a deep learning approach based on graphs to detect plantation lines in UAV-based RGB imagery, presenting a challenging scenario containing spaced plants. The first module of our method extracts a feature map throughout the backbone, which consists of the initial layers of the VGG16. This feature map is used as an input to the Knowledge Estimation Module (KEM), organized in three concatenated branches for detecting 1) the plant positions, 2) the plantation lines, and 3) the displacement vectors between the plants. A graph modeling is applied considering each plant position on the image as vertices, and edges are formed between two vertices (i.e. plants). Finally, the edge is classified as pertaining to a certain plantation line based on three probabilities (higher than 0.5): i) in visual features obtained from the backbone; ii) a chance that the edge pixels belong to a line, from the KEM step; and iii) an alignment of the displacement vectors with the edge, also from the KEM step. Experiments were conducted initially in corn plantations with different growth stages and patterns with aerial RGB imagery to present the advantages of adopting each module. We assessed the generalization capability in the other two cultures (orange and eucalyptus) datasets. The proposed method was compared against state-of-the-art deep learning methods and achieved superior performance with a significant margin considering all three datasets. This approach is useful in extracting lines with spaced plantation patterns and could be implemented in scenarios where plantation gaps occur, generating lines with few-to-no interruptions.

HiDiffSeg: A hierarchical diffusion model for blood vessel segmentation in retinal fundus images

In medical diagnosis, achieving accurate segmentation of fundus vessels is crucial for understanding and identifying various anatomical structures. However, traditional algorithms often face challenges in achieving precise segmentation due to the poor quality of fundus images and the complex branching structure of vessels. To address this issue, we propose a novel approach called HiDiffSeg, which is a coarse-to-fine hierarchical diffusion model for blood vessel segmentation. Our method integrates an enhanced processing coarse-to-fine strategy for retinal fundus images into the denoising process of the diffusion model. The diffusion model is divided into two modules: the dual-guidance module (DGM) and the refinement-guidance module (RGM). The DGM takes the vascular skeleton image and the initial denoised vessel segmentation image generated by the vascular image enhancement model as conditions. Meanwhile, the RGM uses the fundus image as a condition to obtain more accurate results through iterative refinement. To emphasize the significance of vessel edges, we introduce a vessel enhancement module. By taking the original image as input, we generate a pixel-wise edge attention map, assigning greater importance to corresponding edge pixels. To the best of our knowledge, this is the first time a hierarchical diffusion model has been applied to fundus vessel segmentation. We demonstrate the accuracy of our method on three publicly available fundus retinal datasets (i.e., DRIVE, STARE, and CHASE_DB1) using evaluation metrics and compare it with eleven state-of-the-art fundus vessel segmentation methods.

Pixelwise immersion factor calibration for underwater hyperspectral imaging instruments

In situ spectral reflectance initially captured at high spatial resolution with underwater hyperspectral imaging (UHI) is effective for classification and quantification in oceanic biogeochemical studies; however, the measured spectral radiance is rarely used as an absolute quantity due to challenges in calibration of UHI instruments. In this paper, a commercial UHI instrument was calibrated for radiometric flat field response and pixelwise immersion effect to support in situ measurement of absolute spectral radiance. The radiometric and immersion factor calibrations of the UHI instrument were evaluated quantitatively through comparative experiments with a spectroradiometer and a spectrometer. Results show that the immersion factor of the center pixel of the tested UHI instrument was 1.763 in pure water at 600 nm, and the averaged difference in immersion factor between the center and edge pixel of the UHI instrument in the visible light band was only 1∼3% across its half angle field of view of 35° in air. The new calibration coefficients were further used to calculate the spectral radiance of transmitted sunlight through ice algae clusters in sea ice measured by the UHI instrument during an Arctic under-ice bio-optical survey.

Design and Performance of a 30/40 GHz Diplexed Focal Plane for the BICEP Array

We demonstrate a wideband diplexed focal plane suitable for observing low-frequency foregrounds that are important for cosmic microwave background polarimetry. The antenna elements are composed of slotted bowtie antennas with 60% bandwidth that can be partitioned into two bands. Each pixel is composed of two interleaved 12 × 12 pairs of linearly polarized antenna elements forming a phased array, designed to synthesize a symmetric beam with no need for focusing optics. The signal from each antenna element is captured in-phase and uniformly weighted by a microstrip summing tree. The antenna signal is diplexed into two bands through the use of two complementary, six-pole Butterworth filters. This filter architecture ensures a contiguous impedance match at all frequencies, and thereby achieves minimal reflection loss between both bands. Subsequently, out-of-band rejection is increased with a bandpass filter and the signal is then deposited on a transition-edge sensor bolometer island. We demonstrate the performance of this focal plane with two distinct bands, 30 and 40 GHz, each with a bandwidth of ∼20 and 15 GHz, respectively. The unequal bandwidths between the two bands are caused by an unintentional shift in diplexer frequency from its design values. The end-to-end optical efficiency of these detectors is relatively modest, at 20%–30%, with an efficiency loss due to an unknown impedance mismatch in the summing tree. Far-field beam maps show good optical characteristics, with edge pixels having no more than ∼5% ellipticity and ∼10%–15% peak-to-peak differences for A–B polarization pairs.

Optimized multi-scale framework for image enhancement using spatial information-based histogram equalization

ABSTRACT The histogram equalization method, used to improve images, minimizes the amount of pixel intensities, which results in the loss of detail and an unnatural appearance. This study presents an approach for enhancing low contrast images based on their inherent characteristics. The statistical parameter skewness is derived from the photographs to facilitate the classification process into dark and bright images. Based on the classification, appropriate dark and bright pass filters are applied on the multi-level decomposed images to extract the significant edge details. The level of decomposition is optimized using particle swarm optimization. The extracted edge details are utilized by the two-dimensional histogram equalization technique. It leverages the combined presence of edge information and pixel intensities in the low contrast image. The algorithm's efficacy is assessed on three databases, namely CCID, LOL, and DRESDEN, through the utilization of standard deviation (SD), contrast improvement index (CII), discrete entropy (DE), the natural image quality evaluator (NIQE), and Kullback–Leibler distance (KL). Based on the empirical findings, it can be observed that the suggested methodology exhibits better performance compared to alternative methods, including deep learning architectures, in terms of high CII, SD, DE, and low NIQE, KL values.

Projection domain decomposition denoising algorithm based on low rank and similarity-based regularization.

Projection Domain Decomposition (PDD) is a dual energy reconstruction method which implements the decomposition process before image reconstruction. The advantage of PDD is that it can alleviate beam hardening artifacts and metal artifacts effectively as energy spectra estimation is considered in PDD. However, noise amplification occurs during the decomposition process, which significantly impacts the accuracy of effective atomic number and electron density. Therefore, effective noise reduction techniques are required in PDD. This study aims to develop a new algorithm capable of minimizing noise while simultaneously preserving edges and fine details. In this study, a denoising algorithm based on low rank and similarity-based regularization (LRSBR) is presented. This algorithm incorporates the low-rank characteristic of tensors into similarity-based regularization (SBR) framework. This method effectively addresses the issue of instability in edge pixels within the SBR algorithm and enhances the structural consistency of dual-energy images. A series of simulation and practical experiments were conducted on a dual-layer dual-energy CT system. Experiments demonstrate that the proposed method outperforms exiting noise removal methods in Peak Signal-to-noise Ratio (PSNR), Root Mean Square Error (RMSE), and Structural Similarity (SSIM). Meanwhile, there has been a notable enhancement in the visual quality of CT images. The proposed algorithm has a significantly improved noise reduction compared to other competing approach in dual-energy CT. Meanwhile, the LRSBR method exhibits outstanding performance in preserving edges and fine structures, making it practical for PDD applications.

Ultra-large scale array silicon pixel sensors with uniform and low leakage current for advanced X-ray light sources

The development of silicon pixel sensors (SPS) with high operating voltage, low leakage currents, and large arrays can contribute to improving the energy and spatial resolution of advanced X-ray light source detection systems. The Future Detection System comprises a hybrid-pixel detector with a collective resolution of 2048 × 2048 pixels, each measuring 100 μm× 100 μm. It consists of 16 p-i-n SPSs, where each sensor has an array size of 1024× 256 pixels. In this paper, the design of the pixel and guard rings is optimized to achieve uniform and ultra-low pixels leakage currents under high operating voltage. The high leakage current uniformity of the designed sensor is demonstrated through several tests conducted on small scale array SPS. The leakage current of the tested pixels is in the range of 0.50–0.55 pA at room temperature with less than 5% leakage deviation on the whole array. It is accompanied by breakdown voltages greater than 1000 V. The optimized 256× 128 pixel SPS showcases uniform leakage currents below 0.6 pA per pixel at room temperature, as evidence in both the edge and central pixels. The 1024× 256 pixels SPS is then manufactured based on the optimized design results. The obtained results show that the breakdown voltage is greater than 1000 V and the leakage current of the pixel is less than 2.5 pA. In addition, the interpixel capacitance of the sensor also reach an ultra-low level of 16 fF. This study paves the way for the development of a robust semiconductor device solution for applications where ultra-fast and large panel-pixel detectors in advanced X-ray light source detection systems are required.

Evaluating the spatial representativeness of ground-based observations for satellite total ozone products

External verification based on ground-based data is commonly used to assess the accuracy of satellite products for total ozone. However, the spatial representativeness of ground-based observations at the pixel scale and the effects of the spatial mismatch between ground observations and satellite pixels are often overlooked. Therefore, this study focused on an Ozone Monitoring Instrument (OMI) and proposed a novel method to evaluate spatial representativeness using the total ozone product itself without relying on higher spatial resolution imagery from other sensors. A Pixel-level Spatial Mismatch Index (PSMI) was developed to quantify the peaks and stable intervals of the spatial response to the relative deviation of total ozone observations. The analysis revealed that increased uncertainty at the pixel edges can cause significant relative deviation, with the peak deviation occurring within a range of 4–7 km near the pixel boundary. The position of the high precision interval in PSMI was found to be <–22.38 km or >10 km, with mean values of 3.70 % and standard deviations of 3.45 % for the relative deviation. This interval is spatially distance-independent and can be observed on both sides of the pixel boundary. These findings highlight the importance of considering spatial patterns between satellite pixels and ground stations for external validation. Furthermore, avoiding a spatial response peak of the relative deviation at the pixel edge is crucial. Overall, this research provides new perspectives and insights for the verification of total ozone products. By considering these factors mentioned above, future verification of total ozone products can be improved, leading to enhanced their evaluation systems.

Two-Stage Method for Clothing Feature Detection

The rapid expansion of e-commerce, particularly in the clothing sector, has led to a significant demand for an effective clothing industry. This study presents a novel two-stage image recognition method. Our approach distinctively combines human keypoint detection, object detection, and classification methods into a two-stage structure. Initially, we utilize open-source libraries, namely OpenPose and Dlib, for accurate human keypoint detection, followed by a custom cropping logic for extracting body part boxes. In the second stage, we employ a blend of Harris Corner, Canny Edge, and skin pixel detection integrated with VGG16 and support vector machine (SVM) models. This configuration allows the bounding boxes to identify ten unique attributes, encompassing facial features and detailed aspects of clothing. Conclusively, the experiment yielded an overall recognition accuracy of 81.4% for tops and 85.72% for bottoms, highlighting the efficacy of the applied methodologies in garment categorization.

Edge Detection on Field Programmable Gate Array (FPGA)

Edges are significant local changes in the image and are very important features for analysing images. Edges typically occur on the boundary between two different regions in an image. An edge in an image is a significant local change in the image intensity, usually associated with a discontinuity in either the image intensity or the first derivative of the image intensity. Edge detection is one of the tools that can be used in many image processing applications for withdrawing information from image. Sobel edge detection is one of the operators that operate on gradient based edge selection methods to find edge pixels in the image. This report proposed an implementation of Sobel edge Detection algorithm to find the pixels in grayscale images. We present a video streaming architecture and IP implementation using high level synthesis. A video with 720p resolution streamed from the HDMI source and real time edge detected video captured on another monitor. For implementation, ZYBO board or kit is used with Vivado software with 2018.3 V, which has provided adequate peripherals for implementation.

Convolutional transformer attention network with few-shot learning for grassland degradation monitoring using UAV hyperspectral imagery

ABSTRACT In recent years, the desertification of grasslands has increased due to various factors, including both global warming and human activities. It is an essential basis for grassland degradation monitoring to monitor the dynamic change of desert grassland vegetation communities and distribution statistics. Although unmanned aerial vehicle (UAV) remote sensing images have allowed us to achieve dynamic real-time grassland monitoring, the distribution of desert grassland ground objects can be random and narrow, thus increasing the difficulty of sample labelling of remote sensing imagery. Therefore, to reduce the number of samples required for the model, this research proposes a convolutional transformer attention network (CTAN) to identify desert grassland ground objects and validate it on a self-collected desert grassland dataset. The network utilizes the transformer model to enhance its learning of global pixels so that it suppresses the transmission of background pixels within the network. Furthermore, the edge convolution module is designed to strengthen the network’s learning for edge pixels, improving its identification effect. The results show that the network provides 97.22% of overall accuracy (OA), 94.35% of average accuracy (AA), and 0.9398 of Kappa for ground object recognition in desert grassland. The model improves OA by 2.36–9.85% points compared to methods in the same field and 0.8–6.35% points compared to methods in hyperspectral imagery classification. The experimental results show the superior performance of the CTAN model for recognizing desert grassland objects, which helps the management and restoration of desert grasslands.

METHODOLOGY FOR IMPROVING DEEP LEARNING-BASED CLASSIFICATION FOR CT SCAN COVID-19 IMAGES

Early identification of COVID-19 necessitates a precise interpretation of computed tomography (CT) chest images. Consistent and accurate photographs are critical for the correct diagnosis. In this paper, a novel pre-processing approach (mainly improves the contrast) based on gradient enhancement method (GCE) is proposed for prominent visualization of the diagnostic features from the COVID-19 CT images. This pre-processing stage helps in preserving the diagnostic information in the disease affected area. Edge information in the CT images helps the physicians and classification model for better classification. The edge features are preserved by improving the contrast by using multi-scale dependent dark pass filter. From the edge features and pixel intensities, cumulative distribution function (CDF) is computed. It is then mapped to a uniform distribution, resulting in contrast enhanced COVID-19 CT images. These pre-processed images are fed to the customized deep convolutional neural networks (CNNs) like AlexNet, VGG-19, ResNet-101, DenseNet-201, GoogleNet, MobileNet-v2, SqueezeNet, Inception-v3, Xception, and EfficientNet-b0 for classification. Introducing GCE as a pre-processing stage improves the COVID-19 classification accuracy by nearly 6%. Evaluation of the contrast enhancement by GCE technique is carried out on the CT images by contrast improvement index (CII), discrete entropy (DE), and Kullback–Leibler distance (KL-Distance) measures. The experimental findings reveal that the GCE method produces higher CII and DE values than the other enhancement methods available.