Multi-scale Retinex Algorithm Research Articles

Adaptive classification of artistic images using multi-scale convolutional neural networks

The current art image classification methods have low recall and accuracy rate issues . To improve the classification performance of art images, a new adaptive classification method is designed employing multi-scale convolutional neural networks (CNNs). Firstly, the multi-scale Retinex algorithm with color recovery is used to complete the enhancement processing of art images. Then the extreme pixel ratio is utilized to evaluate the image quality and obtain the art image that can be analyzed. Afterward, edge detection technology is implemented to extract the key features in the image and use them as initial values of the item to be trained in the classification model. Finally, a multi-scale convolutional neural network (CNN) is constructed by using extended convolutions, and the characteristics of each level network are set. The decision fusion method based on maximum output probability is employed to calculate different subclassifies’ probabilities and determine the final category of an input image to realize the art image adaptive classification. The experimental results show that the proposed method can effectively improve the recall rate and precision rate of art images and obtain reliable image classification results.

A novel multiscale cGAN approach for enhanced salient object detection in single haze images

In computer vision, image dehazing is a low-level task that employs algorithms to analyze and remove haze from images, resulting in haze-free visuals. The aim of Salient Object Detection (SOD) is to locate the most visually prominent areas in images. However, most SOD techniques applied to visible images struggle in complex scenarios characterized by similarities between the foreground and background, cluttered backgrounds, adverse weather conditions, and low lighting. Identifying objects in hazy images is challenging due to the degradation of visibility caused by atmospheric conditions, leading to diminished visibility and reduced contrast. This paper introduces an innovative approach called Dehaze-SOD, a unique integrated model that addresses two vital tasks: dehazing and salient object detection. The key novelty of Dehaze-SOD lies in its dual functionality, seamlessly integrating dehazing and salient object identification into a unified framework. This is achieved using a conditional Generative Adversarial Network (cGAN) comprising two distinct subnetworks: one for image dehazing and another for salient object detection. The first module, designed with residual blocks, Dark Channel Prior (DCP), total variation, and the multiscale Retinex algorithm, processes the input hazy images. The second module employs an enhanced EfficientNet architecture with added attention mechanisms and pixel-wise refinement to further improve the dehazing process. The outputs from these subnetworks are combined to produce dehazed images, which are then fed into our proposed encoder–decoder framework for salient object detection. The cGAN is trained with two modules working together: the generator aims to produce haze-free images, whereas the discriminator distinguishes between the generated haze-free images and real haze-free images. Dehaze-SOD demonstrates superior performance compared to state-of-the-art dehazing methods in terms of color fidelity, visibility enhancement, and haze removal. The proposed method effectively produces high-quality, haze-free images from various hazy inputs and accurately detects salient objects within them. This makes Dehaze-SOD a promising tool for improving salient object detection in challenging hazy conditions. The effectiveness of our approach has been validated using benchmark evaluation metrics such as mean absolute error (MAE), peak signal-to-noise ratio (PSNR), and structural similarity index measure (SSIM).

Research on the training method of special strength quality of competitive taekwondo based on multi-scale Retinex algorithm under the background of “Internet+”

Research on the training method of special strength quality of competitive taekwondo based on multi-scale Retinex algorithm under the background of “Internet+”

A Method for Electric Vehicle Charging Port Recognition in Complex Environments based on Wavelet Enhancement and Improved Canny Algorithm

Background: The current method has low detection accuracy for electric vehicle charging ports in complex environments. background: The current method has low detection accuracy for electric vehicle charging ports in complex environments. Objective: This paper proposes a method for electric vehicle charging port recognition based on wavelet enhancement and an improved canny algorithm. objective: This paper proposes a method for electric vehicle charging port recognition based on wavelet enhancement and improved canny algorithm. Methods: Firstly, a preprocessing assessment model is proposed to determine whether preprocessing enhancement should be applied. Subsequently, we perform a stretch transformation on the S channel in the HSI color space and apply a discrete wavelet transform to the I channel to restore the brightness and detail features of the target objects in the image. Specifically, in this paper, adaptive median filtering and multi-scale Retinex algorithm are utilized on different frequency components respectively. Finally, the features and parameters of the charging port are detected and extracted through the maximum inter-class variance method, morphological processing, the improved Canny algorithm, and Hough transform fitting. Results: This paper conducts comparative experiments on preprocessing enhancement of charging port images in different environments and features detection of charging ports. The hole feature parameters detected by this method are consistent with the actual situation. Conclusion: The experiment proves that the method proposed in this paper can meet the requirements of precise identification of charging ports in complex environments.

Transmission line tower inclination measurement method based on three-dimensional laser scanning and inter frame difference

Abstract The purpose of this study is to propose a method based on three-dimensional laser scanning and frame difference technology to measure the inclination of transmission line towers, so as to realize accurate monitoring of transmission line structure. Firstly, the multi-scale Retinex algorithm is used to enhance and preprocess the transmission line tower image to improve the image quality. Secondly, the method of frame difference and background difference is combined to detect and identify the inclined tower. Finally, the point cloud data of inclined tower is obtained by using three-dimensional laser scanning technology, and the inclination of transmission line tower is calculated by coordinates. The experimental results show that this method has high accuracy and reliability, with an error of less than 0.15 × 10−5 and an inclination of less than 0.17°. It is proved that this method can improve the monitoring efficiency of transmission lines, has good effects in engineering practice, and can provide effective means for the safe operation of transmission line structures.

Low-Light Mine Image Enhancement Algorithm Based on Improved Retinex

Aiming at solving the problems of local halo blurring, insufficient edge detail preservation, and serious noise in traditional image enhancement algorithms, an improved Retinex algorithm for low-light mine image enhancement is proposed. Firstly, in HSV color space, the hue component remains unmodified, and the improved multi-scale guided filtering and Retinex algorithm are combined to estimate the illumination and reflection components from the brightness component. Secondly, the illumination component is equalized using the Weber–Fechner law, and the contrast limited adaptive histogram equalization (CLAHE) is fused with the improved guided filtering for the brightness enhancement and denoising of reflection component. Then, the saturation component is adaptively stretched. Finally, it is converted back to RGB space to obtain the enhanced image. By comparing with single-scale Retinex (SSR) algorithm and multi-scale Retinex (MSR) algorithm, the mean, standard deviation, information entropy, average gradient, peak signal-to-noise ratio (PSNR), and structural similarity (SSIM) are improved by an average of 50.55%, 19.32%, 3.08%, 28.34%, 29.10%, and 22.97%. The experimental dates demonstrate that the algorithm improves image brightness, prevents halo artifacts while retaining edge details, reduces the effect of noise, and provides some theoretical references for low-light image enhancement.

Image Enhancement Method for Photoacoustic Imaging of Deep Brain Tissue

Photoacoustic imaging (PAI) is an emerging biomedical imaging modality, offering numerous advantages, including high resolution and high contrast. In its application to brain imaging, however, the photoacoustic (PA) signals from brain tissue weaken considerably due to the distortion effects of the skull. This attenuation reduces the resolution and contrast significantly. To address this issue, here we describe a Log-MSR algorithm that combines the logarithmic depth logarithmic enhancement (Log) algorithm and the multi-scale Retinex (MSR) algorithm. In this method, the Log algorithm performs local weighted compensation based on signal attenuation for different depths, while the MSR algorithm improves the contrast of the image. The proposed Log-MSR algorithm was tested and validated using several phantom and in vivo experiments. The enhanced images constructed by the Log-MSR algorithm were qualitatively and quantitatively analyzed in terms of brain structure and function. Our results show that the Log-MSR algorithm may provide a significant enhancement to photoacoustic imaging of deep brain tissue.

Research based on the use of digital media art in garden landscape design

Abstract This paper takes garden landscape as the research object, discusses the application of digital media art in assisting garden landscape design, and obtains the optimization probability of various types of garden landscape types in landscape spatial layout through patch generation using the change simulation (PLUS) model. And after using the multi-scale Retinex algorithm to make image enhancement, a fusion of single-scale enhancement results in achieving the initial optimization of the image, after a dynamic interception and stretching operation to restore the enhancement effect to achieve the optimization of the landscape image. Finally, a group of landscape design images are selected as experimental objects to test the effectiveness of digital media art-assisted landscape design. The results show that with the assistance of digital media art, the modulus of the change distance of neighboring units is between 2.21 and 10.89, and the relative change rate takes the value between 0.45 and 5.21. The method is capable of balancing the ups and downs and repetitive rhythms in the design, ensuring that the landscape has good brightness and visual effects.

Multi-target tracking algorithm in aquaculture monitoring based on deep learning

In order to analyze fish behavior, real-time underwater fish monitoring is essential. This work presents an underwater multi-target tracking algorithm for aquaculture monitoring, utilizing deep learning techniques. Underwater images are acquired using an underwater robot, and the images are defogged using the multi-scale Retinex algorithm based on HSV space. To enhance object detection performance, GhostNetv2 was integrated into the You Only Look Once version 5 (YOLOv5) object detection algorithm, supplemented by the addition of the Coordinate Attention (CA) module, resulting in the development of GN-YOLOv5. For the tracking algorithm, the more accurate Generalized Intersection over Union (GIoU) method was incorporated into the StrongSORT tracking algorithm. Moreover, to achieve more precise target tracking, a fish re-identification model was established. The proposed algorithms were evaluated through experiments conducted on various datasets, including VOC 2012, MOT16, and self-built datasets. The results demonstrate notable improvements: the GN-YOLOv5 model showed a 32.91% reduction in parameters and a 3.22% increase in precision. Furthermore, the enhanced StrongSORT algorithm exhibited a 3.84% increase in MOTA, a 28.00% reduction in IDS, and a speed boost of 7 FPS, leading to stable and accurate multi-target fish tracking.

Wavelet energy-based adaptive retinex algorithm for low light mobile video enhancement

ABSTRACT Our paper presents an adaptive multiscale retinex algorithm and a new wavelet energy metric to improve low-light video captured on mobile devices. Initially, we extract RGB frames from the video and convert them to hue-saturation-value (HSV) format, preserving the hue channel to prevent common RGB colour shifting issues. Saturation channel enhancement is achieved through histogram equalization (HE), extending the dynamic range. The adaptive retinex algorithm enhances the value channel, quantified by our new wavelet energy metric. Combining the modified value and saturation channels improves the contrast of the reconstructed image. As a final step, we transform the HSV video back to RGB and restore naturalness using a modified colour restoration technique. The proposed approach has been tested on over 300 images and videos. It is evident from the experimental results presented that the proposed method lowers noise and halo artifacts more effectively than existing methods.

Data collection and information security analysis in sports teaching system based on intelligent sensor

With the rapid development of intelligent sensor technology and the application of scenarios, the video images of moving targets collected by intelligent terminal devices can be obtained in sports teaching scenarios, and then through preprocessing, target tracking and behavior analysis, the analysis report is finally generated to achieve the effect of sports training. However, in the process of processing, the edge information of the video image is damaged, the color deviation exists, and when tracking the moving target, the final result will be affected by the deformation and partial occlusion of the target. In view of the above problems, the original bilateral filtering algorithm is changed to the multi-scale Retinex algorithm in the preprocessing to ensure that the video image is clearer and more complete and can be effectively protected. At the same time, for the target tracking algorithm, the Camshift algorithm which takes the color histogram as the tracking target and records the distribution probability is used. So that the results will not be affected by the deformation and partial occlusion of the target. The final experimental results show that the improved pre-processing algorithm and target tracking algorithm have smaller errors in operation, better protection of the edge information of video images, reduced the time consumption of the target tracking unit to be detected, and also improved its anti-jamming stability. In the future development, the image processing algorithm should have a better operation path, further reduce the running time, improve the value of UCIQE and peak signal to noise ratio, and track multiple targets at the same time with the smallest error, capture the moment to reach the single frame level, and also become better for teaching. In the system design, information security performance is compared with vulnerability identification to evaluate the security level of physical education teaching system.

Infrared image enhancement algorithm based on improved wavelet threshold function and weighted guided filtering

Aiming at the problems such as the poor visual effect of infrared thermal imaging under low illumination conditions, an enhanced infrared image recognition method based on wavelet decomposition is proposed for this paper. First, according to wavelet decomposition, the infrared image is divided into a low-frequency sub-band (LFS) and high-frequency sub-band (HFS) in each orientation, and HFS in each orientation is denoised using a modified wavelet threshold function to strengthen the denoising effect of the infrared image; Second, the multi-scale Retinex (MSR) algorithm based on weighted guided filtering (WGIF) is used to evaluate the component of illumination in the basic layer of the LFS, and the WGIF-segmented detail layer images are fused with the MSR processed ones to effectively highlight the texture details of the LFS; Finally, the LFS and the HFS are wavelets reconstructed (WR) to obtain the infrared enhanced images. This paper applies the algorithm to a low-resolution (resolution 32x32 pixels) infrared thermal imaging module, and the results of combining subjective and objective evaluation indexes show that the overall property index of the algorithm in this paper is superior to other contrasting algorithms and can effectively improve the infrared image quality.

The impact of variable illumination on vegetation indices and evaluation of illumination correction methods on chlorophyll content estimation using UAV imagery

BackgroundThe advancements in unmanned aerial vehicle (UAV) technology have recently emerged as an effective, cost-efficient, and versatile solution for monitoring crop growth with high spatial and temporal precision. This monitoring is usually achieved through the computation of vegetation indices (VIs) from agricultural lands. The VIs are based on the incoming radiance to the camera, which is affected when there is a change in the scene illumination. Such a change will cause a change in the VIs and subsequent measures, e.g., the VI-based chlorophyll-content estimation. In an ideal situation, the results from VIs should be free from the impact of scene illumination and should reflect the true state of the crop’s condition. In this paper, we evaluate the performance of various VIs computed on images taken under sunny, overcast and partially cloudy days. To improve the invariance to the scene illumination, we furthermore evaluated the use of the empirical line method (ELM), which calibrates the drone images using reference panels, and the multi-scale Retinex algorithm, which performs an online calibration based on color constancy. For the assessment, we used the VIs to predict leaf chlorophyll content, which we then compared to field measurements.ResultsThe results show that the ELM worked well when the imaging conditions during the flight were stable but its performance degraded under variable illumination on a partially cloudy day. For leaf chlorophyll content estimation, The r^2 of the multivariant linear model built by VIs were 0.6 and 0.56 for sunny and overcast illumination conditions, respectively. The performance of the ELM-corrected model maintained stability and increased repeatability compared to non-corrected data. The Retinex algorithm effectively dealt with the variable illumination, outperforming the other methods in the estimation of chlorophyll content. The r^2 of the multivariable linear model based on illumination-corrected consistent VIs was 0.61 under the variable illumination condition.ConclusionsOur work indicated the significance of illumination correction in improving the performance of VIs and VI-based estimation of chlorophyll content, particularly in the presence of fluctuating illumination conditions.

A novel coal-gangue recognition method in underground coal mine based on image processing

ABSTRACT This study proposes a coal-gangue recognition method based on Retinex and extreme learning machine. The traditional Retinex has many limitations, including halo artifacts, excessive enhancement, and noise amplification in dark areas. To handle these, first, a modified multi-scale Retinex image enhancement algorithm based on hue, saturation, value color space, and bidimensional empirical mode decomposition named modified multi-scale Retinex algorithm with color restoration (MMSRCR) is proposed herein. On-site images are used to verify the effectiveness of the MMSRCR. Second, a preliminary segmentation method for coal-gangue images based on the optimal entropy threshold optimized by the improved quantum genetic algorithm is designed. The proposed segmentation method performs excellent stability compared with other traditional methods. Third, the adhesion part of the ore is further segmented based on the morphology opening operation and watershed algorithm. Then, the feature space of each enclosed area is extracted using the chain code and line segment tables. The improved extreme learning machine completes the classification of these areas, and the coal-gangue visual ratio recognition method is proposed. Finally, an experimental platform for visual ratio recognition is built to verify the proposed method. The experimental results show that the proposed method can accurately identify the visual ratio of coal-gangue.

Fire Sensor and Surveillance Camera-Based GTCNN for Fire Detection System

Fire accident is a disaster that can happen anytime anywhere due to accidental causes. In existing works, sensor- and computer vision-based approaches have been used for developing the fire detection model, but they fail to attain the accurate results. The sensor-based methods need more time to detect the fire locations and detection coverage also less. The camera sometimes will consider heavy sunlight as fire and it leads to false positive result, which degrades the accuracy. To overcome the above problems, in this research, a novel optimized Gaussian probability-based threshold convolutional neural network (GTCNN) model has been proposed for detecting the fire accidents using various sensors and surveillance camera-based video (SV). Sensor features map has been calculated from various fire sensors and frames/images from SV are preprocessed using a multiscale retinex algorithm. In addition, the Gaussian threshold (GT) logically integrates with the feature map to increase fire pixel count in low-resolution images. The probability results from sensors and SV camera are optimized by multiobjective mayfly optimization (MOMO) algorithm that normalizes the network parameters, which gives the accurate result. The performance of the proposed optimized GTCNN net is different from the existing deep learning networks in terms of multifeature processing. The result of the proposed work attains the detection accuracy of 98.23%. The proposed optimized GTCNN improves the overall accuracy of 3.25%, 3.79%, and 0.21% better than the channel attention mechanism, lightweight CNN, and you only look once (YOLOv5m), respectively.

Research on Improved Retinex-Based Image Enhancement Method for Mine Monitoring

An improved Retinex fusion image enhancement algorithm is proposed for the traditional image denoising methods and problems of halo enlargement and image overexposure after image enhancement caused by the existing Retinex algorithm. First, a homomorphic filtering algorithm is used to enhance each RGB component of the underground coal mine surveillance image and convert the image from RGB space to HSV space. Second, bilateral filtering and multi-scale retinex with color restoration (MSRCR) fusion algorithms are used to enhance the luminance V component while keeping the hue H component unchanged. Third, adaptive nonlinear stretching transform is used for the saturation S-component. Last, the three elements are combined and converted back to RGB space. MATLAB simulation experiments verify the superiority of the improved algorithm. Based on the same dataset and experimental environment, the improved algorithm has a more uniform histogram distribution than the multi-scale Retinex (msr) algorithm and MSRCR algorithm through comparative experiments. At the same time, the peak signal-to-noise ratio (PSNR), structural similarity (SSIM), standard deviation, average gradient, mean value, and colour picture information entropy of the images were improved by 8.28, 0.15, 4.39, 7.38, 52.92 and 2.04, respectively, compared to the MSR algorithm, and 3.97, 0.02, 34.33, 60.46, 26.21, and 1.33, respectively, compared to the MSRCR algorithm. The experimental results show that the image quality, brightness and contrast of the images enhanced by the improved Retinex algorithm are significantly enhanced, and the amount of information in the photos increases, the halo and overexposure in the images are considerably reduced, and the anti-distortion performance is also improved.

Double-function enhancement algorithm for low-illumination images based on retinex theory.

In order to solve the problems of noise amplification and excessive enhancement caused by low contrast and uneven illumination in the process of low-illumination image enhancement, a high-quality image enhancement algorithm is proposed in this paper. First, the total-variation model is used to obtain the smoothed V- and S-channel images, and the adaptive gamma transform is used to enhance the details of the smoothed V-channel image. Then, on this basis, the improved multi-scale retinex algorithms based on the logarithmic function and on the hyperbolic tangent function, respectively, are used to obtain different V-channel enhanced images, and the two images are fused according to the local intensity amplitude of the images. Finally, the three-dimensional gamma function is used to correct the fused image, and then adjust the image saturation. A lightness-order-error (LOE) approach is used to measure the naturalness of the enhanced image. The experimental results show that compared with other classical algorithms, the LOE value of the proposed algorithm can be reduced by 79.95% at most. Compared with other state-of-the-art algorithms, the LOE value can be reduced by 53.43% at most. Compared with some algorithms based on deep learning, the LOE value can be reduced by 52.13% at most. The algorithm proposed in this paper can effectively reduce image noise, retain image details, avoid excessive image enhancement, and obtain a better visual effect while ensuring the enhancement effect.

Autonomous recognition and positioning of shield segments based on red, green, blue and depth information

Automatic shield segment assembly can reduce intensive labour and increase tunnel boring efficiency. However, insufficient lighting and multi-interference environments significantly affect segment recognition and positioning accuracy. This study develops a highly reliable and low-cost recognition and positioning system, including both hardware and software, for segment markers based on real-time monocular vision and light-insensitive depth measurements. An improved discrete wavelet transform based on the HSV channels and an adaptive multiscale Retinex algorithm with colour recovery are developed to enhance image brightness and colour contrast. Compared to traditional feature matching, support vector machine recognition and classification increase the real-time recognition accuracy. One hundred full-scale and field tests were performed under the original, insufficient, and interferential lighting environments. The recognition time was approximately 0.7 s, and the recognition accuracy was higher than 90%. A grasping experiment verified the feasibility of the system.

Terahertz image enhancing based on the physical model and multiscale retinex algorithm.

To settle the terahertz (THz) image degradation problem, we propose an effective enhancement method based on the physical model and multiscale retinex (MSR) algorithm. The overall enhancing process involves two parts: reconstruction and enhancement. First, the original THz images are reconstructed by a mathematical model, which is built and considered the THz absorption variate and Gaussian distribution of the beam. Then, the original images are processed by the proposed algorithm, which combines the atmospheric scattering model and optimized MSR algorithm. The proposed algorithm not only recovers the image scene radiance and removes haze, but also can make a compromise of the dynamic range of gray scale and edge enhancement of the image. Results on a variety of THz images demonstrate our method can effectively improve the quality of THz images and retain sufficient image details.

Recognition of road cracks based on multi-scale Retinex fused with wavelet transform

Cracks are the main diseases of roads and potential threats to road safety. The detection and repair of cracks is the focus of intelligent transportation system research. However, the performance of automatic crack detection is often not good enough due to the uneven illumination, low contrast between the crack and the surrounding pavement and the possible presence of shadows similar in intensity to the crack. In this paper, an improved multi-scale Retinex algorithm is proposed to enhance the crack image. The wavelet transform is integrated into the traditional multi-scale Retinex algorithm to avoid the halo generated by the Retinex algorithm, thereby reducing the image distortion. Meanwhile, the multi-scale Retinex algorithm can make up for the lack of useful information lost by wavelet transform, so the combination of the two can obtain better crack enhancement effect. In addition, the preprocessing of shadow removal is performed before crack enhancement, which effectively eliminates the interference of high-intensity shadows. Through the comparison of objective performance indicators, the newly proposed algorithm can better highlight the crack information. The method proposed in this paper can effectively realize the functions of shadow removal and crack enhancement, so that the recognition accuracy of the overall detection system reaches 95.8%, indicating that the algorithm has high research significance and engineering application value.