Similar Gray Values Research Articles

Pellet image segmentation model of superpixel feature-based support vector machine in digital twin

A digital twin model based on superpixel features is established to solve the problem of noise and similar gray values between foreground and background of pellet images. With superpixel as the basic unit of segmentation, the influence of single pixel on segmentation results is reduced, and allows for higher segmentation accuracy. The gray-level co-occurrence matrix is used to represent the superpixel characteristic information, and the color moment and gray level distribution are combined to comprehensively characterize the superpixel. Through principal component analysis and correlation analysis, The feature compression of superpixel is realized, and the computational efficiency is improved. The superpixel binary classification data set is built, and the multi-dimensional feature information of superpixel is extracted as input vector to train the binary classification model of SVM, and the image segmentation problem is transformed into foreground and background classification problem. A multi-scale superpixel segmentation boundary optimization method is proposed to further refine the boundary region of foreground and background. A four-neighborhood search algorithm is proposed to reduce the missegmentation rate of edge superpixels. Experimental results show that the accuracy of the proposed method can reach 95.87%, the precision of image edge segmentation is high, and the foreground and background of granular image are accurately segmented. The digital twin model is established which can provide basis for the subsequent visual analysis and decision.

Two-dimensional reciprocal cross entropy multi-threshold combined with improved firefly algorithm for lung parenchyma segmentation of COVID-19 CT image

The lesions of COVID-19 CT image show various kinds of ground-glass opacity and consolidation, which are distributed in left lung, right lung or both lungs. The lung lobes are uneven and it have similar gray value to the surrounding arteries, veins, and bronchi. The lesions of COVID-19 have different sizes and shapes in different periods. Accurate segmentation of lung parenchyma in CT image is a key step in COVID-19 detection and diagnosis. Aiming at the unideal effect of traditional image segmentation methods on lung parenchyma segmentation in CT images, a lung parenchyma segmentation method based on two-dimensional reciprocal cross entropy multi-threshold combined with improved firefly algorithm is proposed. Firstly, the optimal threshold method is used to realize the initial segmentation of the lung, so that the segmentation threshold can change adaptively according to the detailed information of lung lobes, trachea, bronchi and ground-glass opacity. Then the lung parenchyma is further processed to obtain the lung parenchyma template, and then the defective template is repaired combined with the improved Freeman chain code and Bezier curve. Finally, the lung parenchyma is extracted by multiplying the template with the lung CT image. The accuracy of lung parenchyma segmentation has been improved in the contrast clarity of CT image and the consistency of lung parenchyma regional features, with an average segmentation accuracy rate of 97.4%. The experimental results show that for COVID-19 and suspected cases, the method has an ideal segmentation effect, and it has good accuracy and robustness.

Asymptomatic COVID-19 CT image denoising method based on wavelet transform combined with improved PSO

The quality of asymptomatic corona virus disease 2019 (COVID-19) computed tomography (CT) image is reduced due to interference from Gaussian noise, which affects the subsequent image processing. Aiming at the problem that asymptomatic COVID-19 CT image often have small flake ground-glass shadow in the early lesions, and the density is low, which is easily confused with noise. A denoising method of wavelet transform with shrinkage factor is proposed. The threshold decreases with the increase of decomposition scale, and it reduces the misjudgment of signal points. In the advanced stage, the range of lesions increases, with consolidation and fibrosis in different sizes, which have similar gray value to the CT images of suspected cases. Aiming at the problems of low contrast and fuzzy boundary in the traditional wavelet transform, the threshold function based on the optimization of parameters combined with the improved particle swam optimization (PSO) is proposed, so that the parameters of wavelet threshold function can change adaptively according to the lung lobe and ground-glass lesions with fewer iterations. The simulation results show that the paper method is significantly better than other algorithms in peak signal-to-noise ratio (PSNR), signal-to-noise ratio (SNR) and mean absolute error (MSE). For example, aiming at the early asymptomatic COVID-19, compared with the comparison methods, the PSNR under the proposed method has increased by about 5 dB, the MSE has been greatly reduced, and the SNR has increased by about 6.1 dB. It can be seen that the denoising effect under the proposed method is the best.

COVID-19 CT image denoising algorithm based on adaptive threshold and optimized weighted median filter

CT image of COVID-19 is disturbed by impulse noise during transmission and acquisition. Aiming at the problem that the early lesions of COVID-19 are not obvious and the density is low, which is easy to confuse with noise. A median filtering algorithm based on adaptive two-stage threshold is proposed to improve the accuracy for noise detection. In the advanced stage of ground-glass lesion, the density is uneven and the boundary is unclear. It has similar gray value to the CT images of suspected COVID-19 cases such as adenovirus pneumonia and mycoplasma pneumonia (reticular shadow and strip shadow). Aiming at the problem that the traditional weighted median filter has low contrast and fuzzy boundary, an adaptive weighted median filter image denoising method based on hybrid genetic algorithm is proposed. The weighted denoising parameters can adaptively change according to the detailed information of lung lobes and ground-glass lesions, and it can adaptively match the cross and mutation probability of genetic combined with the steady-state regional population density, so as to obtain a more accurate COVID-19 denoised image with relatively few iterations. The simulation results show that the improved algorithm under different density of impulse noise is significantly better than other algorithms in peak signal-to-noise ratio (PSNR), image enhancement factor (IEF) and mean absolute error (MSE). While protecting the details of lesions, it enhances the ability of image denoising.

SAR Image Segmentation Based on Fisher Vector Superpixel Generation and Label Revision

This paper addresses the problem of superpixel-bases segmentation of synthetic aperture radar (SAR) images. Most superpixel segmentation methods have difficulties in segmenting adjacent regions with similar gray values, due to only considering spatial and gray information. To solve this problem and improve segmentation accuracy, this paper proposes a SAR image segmentation method based on Fisher vector superpixel generation and label revision. Firstly, the Fisher vector is obtained by processing the Gaussian mixture function. By introducing the Fisher vector, a distance formula is constructed for superpixel segmentation. Therefore, the adjacent regions with similar gray values can be segmented effectively in the generated superpixel map. Secondly, the superpixels are clustered using the K-means algorithm to obtain the initial label map. Then, with extracted edge information as constraints, the pixel labels obtained by K-means are repaired pixel by pixel to get the middle label map, according to the number and gray value difference of labels. This overcomes influence of noise generated by K-means. Finally, the middle label map is relabeled using the region growth algorithm to divide pixel blocks. Isolated pixel blocks surrounded by similar labels are corrected, based on the gray mean difference. The final label result has a better segmentation accuracy. Experiments on synthetic SAR images and real images demonstrate that the proposed algorithm achieves higher segmentation accuracy than six state-of-the-art clustering algorithms for SAR image segmentation.

GenU-Net++: An Automatic Intracranial Brain Tumors Segmentation Algorithm on 3D Image Series with High Performance

Automatic segmentation of intracranial brain tumors in three-dimensional (3D) image series is critical in screening and diagnosing related diseases. However, there are various challenges in intracranial brain tumor images: (1) Multiple brain tumor categories hold particular pathological features. (2) It is a thorny issue to locate and discern brain tumors from other non-brain regions due to their complicated structure. (3) Traditional segmentation requires a noticeable difference in the brightness of the interest target relative to the background. (4) Brain tumor magnetic resonance images (MRI) have blurred boundaries, similar gray values, and low image contrast. (5) Image information details would be dropped while suppressing noise. Existing methods and algorithms do not perform satisfactorily in overcoming these obstacles mentioned above. Most of them share an inadequate accuracy in brain tumor segmentation. Considering that the image segmentation task is a symmetric process in which downsampling and upsampling are performed sequentially, this paper proposes a segmentation algorithm based on U-Net++, aiming to address the aforementioned problems. This paper uses the BraTS 2018 dataset, which contains MR images of 245 patients. We suggest the generative mask sub-network, which can generate feature maps. This paper also uses the BiCubic interpolation method for upsampling to obtain segmentation results different from U-Net++. Subsequently, pixel-weighted fusion is adopted to fuse the two segmentation results, thereby, improving the robustness and segmentation performance of the model. At the same time, we propose an auto pruning mechanism in terms of the architectural features of U-Net++ itself. This mechanism deactivates the sub-network by zeroing the input. It also automatically prunes GenU-Net++ during the inference process, increasing the inference speed and improving the network performance by preventing overfitting. Our algorithm’s PA, MIoU, P, and R are tested on the validation dataset, reaching 0.9737, 0.9745, 0.9646, and 0.9527, respectively. The experimental results demonstrate that the proposed model outperformed the contrast models. Additionally, we encapsulate the model and develop a corresponding application based on the MacOS platform to make the model further applicable.

Automatic Segmentation of Remote Sensing Images on Water Bodies Based on Image Enhancement

By virtue of high-resolution remote sensing satellites, there is a possibility to analyze remote sensing images on water bodies through digital image processing (DIP). In many remote sensing images, however, the water bodies have similar gray values as other ground objects. To effectively distinguish water bodies from other ground objects in these images, this paper proposes a logarithmic enhancement method for remote sensing images on water bodies based on adaptive morphology. The proposed method can filter the noise of non-target area, and enhance the water body in the original image. On this basis, a morphology-based segmentation method was designed for remote sensing images on water bodies. Experimental results show that our method achieved a high segmentation accuracy, controlling the mean segmentation error at below 1.32%.

A Study on Establishing a Microstructure-Related Hardness Model with Precipitate Segmentation Using Deep Learning Method.

This paper established a microstructure-related hardness model of a polycrystalline Ni-based superalloy GH4720Li, and the sizes and area fractions of γ’ precipitates were extracted from scanning electron microscope (SEM) images using a deep learning method. The common method used to obtain morphological parameters of γ’ precipitates is the thresholding method. However, this method is not suitable for distinguishing different generations of γ’ precipitates with similar gray values in SEM images, which needs many manual interventions. In this paper, we employ SEM with ATLAS (AuTomated Large Area Scanning) module to automatically and quickly detect a much wider range of microstructures. A deep learning method of U-Net is firstly applied to automatically and accurately segment different generations of γ’ precipitates and extract their parameters from the large-area SEM images. Then the obtained sizes and area fractions of γ’ precipitates are used to study the precipitate stability and microstructure-related hardness of GH4720Li alloy at long-term service temperatures. The experimental results show that primary and secondary γ’ precipitates show good stability under long-term service temperatures. Tertiary γ’ precipitates coarsen selectively, and their coarsening behavior can be predicted by the Lifshitz–Slyozov encounter modified (LSEM) model. The hardness decreases as a result of γ’ coarsening. A microstructure-related hardness model for correlating the hardness of the γ’/γ coherent structures and the microstructure is established, which can effectively predict the hardness of the alloy with different microstructures.

Eliminating Lining Seams in Tunnel Concrete Crack Images via Line Segments’ Translation and Expansion

Cracks are the main form of tunnel complications and have drawn much attention in current tunnel safety monitoring. However, automatic crack detection cannot accurately extract cracks from tunnel concrete lining surfaces because of the inference of noises like lining seams, which have similar gray values and textures to cracks, and increase the topological complexity of cracks. In this paper, we proposed a novel method for eliminating lining seams in tunnel concrete crack images, to address the above-mentioned issues. Our contributions are shown as follows: 1) classified arbitrary line segments to better distinguish the edges on the cracks and the seams, and proposed uniform k-divided angle model to get accurate classification and 2) introduced two new principles to help effectively eliminate lining seams through the information of translation direction and expansion width of all line segments. The experimental results proved the superior precision and efficiency of our method compared with the existing methods.

Automatic lung segmentation in the presence of alveolar collapse

Abstract Lung ventilation and perfusion analyses using chest imaging methods require a correct segmentation of the lung to offer anatomical landmarks for the physiological data. An automatic segmentation approach simplifies and accelerates the analysis. However, the segmentation of the lungs has shown to be difficult if collapsed areas are present that tend to share similar gray values with surrounding non-pulmonary tissue. Our goal was to develop an automatic segmentation algorithm that is able to approximate dorsal lung boundaries even if alveolar collapse is present in the dependent lung areas adjacent to the pleura. Computed tomography data acquired in five supine pigs with injured lungs were used for this purpose. First, healthy lung tissue was segmented using a standard 3D region growing algorithm. Further, the bones in the chest wall surrounding the lungs were segmented to find the contact points of ribs and pleura. Artificial boundaries of the dorsal lung were set by spline interpolation through these contact points. Segmentation masks of the entire lung including the collapsed regions were created by combining the splines with the segmentation masks of the healthy lung tissue through multiple morphological operations. The automatically segmented images were then evaluated by comparing them to manual segmentations and determining the Dice similarity coefficients (DSC) as a similarity measure. The developed method was able to accurately segment the lungs including the collapsed regions (DSCs over 0.96).

Robust road detection in shadow conditions

Image-based road detection is a vital task for many real-world applications, such as autonomous driving and obstacle detection. We propose a method for segmenting road regions from a single image based on horizon line estimation and clustering technology. The key idea is to leverage normalized cross correlation to search for the line separating the road image. Additionally, we divide the lower part of the road image into several identical parts horizontally and utilize a density-peak clustering algorithm in terms of gray and HSV value of each pixel. Clustering results are further labeled as road and nonroad based on the assumption that two adjacent horizontal parts share similar clustering size and average gray value. For road images with shadows, we also propose a new shadow-free space derived from HSV space. By calculating maximum entropy and reconstructing images, we show that shadows are eliminated to achieve better road detection. Experimental results on four datasets and shadow images demonstrate the effectiveness and robustness of our method.

Variational Contrast Enhancement of Gray-Scale and RGB Images

The aim of this paper is twofold. First, we propose a new method for enhancing the contrast of gray-value images. We use the difference of the average local contrast measures between the original and the enhanced images within a variational framework. This enables the user to intuitively control the contrast level and the scale of the enhanced details. Moreover, our model avoids large modifications of the original image histogram. Thereby it preserves the global illumination of the scene and it can cope with large areas having similar gray values. The minimizer of the proposed functional is computed by a gradient descent algorithm in connection with a polynomial approximation of the average local contrast measure. The polynomial approximation is computed via Bernstein polynomials. In the second part, the approach is extended to a variational enhancement method for color images. The model approximately preserves the hue of the original image and additionally includes a total variation term to correct the possible noise. The method requires no post- or preprocessing. The minimization problem is solved with a hybrid primal---dual algorithm. Experiments demonstrate the efficiency and the flexibility of the proposed approaches in comparison with state-of-the-art methods.

A method for road extraction from remote sensing imagery

Because the choice of the reasonable and effective threshold of segmentation is rather difficult,the method based on threshold segmentation is not applicable to extracting road information from remote sensing images in that there are obviously multiple types of road and non- road feature interference. To tackle this problem,the authors propose in this paper a method which combines Mean Shift algorithm with threshold segmentation to extract road information. Firstly,the Mean Shift is used to smooth the image,then the texture distribution is made more uniform in the road and the edge of the road is kept. Secondly,the Mean Shift segmentation is used to segment the image,and group the roads which have the same or similar gray values into one gray value showing. Thirdly,as different kinds of color information of the road have different showing characteristics in the gray- level histogram,the gray value which has relatively more numbers of picture elements is taken as the segment range boundary point to obtain the original road information by using multi- threshold segmentation. Finally,post- processing of the original road information is made to obtain the road. The experiment results indicate that this method can extract the road information from the remote sensing imagery and broaden the scope of the use of the threshold segmentation to extract the road information.

River Channel Extraction From SAR Images by Combining Gray and Morphological Features

In remote sensing images acquired by the synthetic aperture radar (SAR), the backscattering coefficient is the function of the gray level. Since the gray of the pixels in the water area is relatively low, the gray level is commonly regarded as a valid feature for river channel identification and extraction. However, the gray threshold method is challenging mostly due to the speckles and false objects with similar gray value to the interest objects. Combining gray and morphological features, a novel multistage method for river channel extraction is proposed in this paper. First, the gray threshold-based image segmentation method is applied for initially removing the background noises. Next, a novel morphological model is proposed for identifying the river channel. After extracting the rough area of the river channel, the gray threshold-based image segmentation is reused for pruning results. Finally, the morphological filter is employed for correcting results. The benefit of the proposed method lies in its ability for properly combining the gray and morphological features, and the advantages of these features are fully exploited. The proposed method was evaluated both objectively and subjectively by utilizing a series of SAR images and quantified criterions. For the overall dataset, the missed detection rate, false alarm rate, and the correctness were statistically calculated, respectively. All experimental results proved that in comparision to the classic Otsu gray threshold method and its updated editions, the proposed approach is fast, robust, and effective for river channel extraction.

Edge Detection Features to Evaluate Hardness of Dates Using Monochrome Images

Date is an important fruit in the regular diets of many peoples in the Arab countries and several other parts of the world. Hardness is one of the important attributes in determining the quality of dates. Hard dates are tough, difficult to chew, unsuitable for several product preparation and ultimately fetching low market price. In general, hard dates have strong curvy and zigzag textured skin. In this study, the efficiency of edge detection features in classifying dates based on hardness using monochrome images was determined. Date samples (Fard variety) were obtained from three major dates growing regions in Oman, and classified into three grades (soft, semi-hard and hard) by a group of trained graders followed with a confirmation by an experienced grader in a commercial dates company. Individual dates were imaged using a monochrome camera (600 dates per grade; total = 1,800 images). A total of 36 features were extracted (28 in spatial domain and 8 in frequency domain) using edge detection methods. An artificial neural network (ANN) was used to classify the dates based on hardness. The overall classification accuracies were 75 % and 87 % while using single ANN (irrespective of regions) for three-class (soft, semi-hard and hard) and two-class (soft and hard (semi-hard and hard together)) models, respectively. While using separate ANN for each region in the three-class model, the mean classification accuracies were 94 %, 59 % and 84 % for soft, semi-hard and hard dates, respectively. Similarly, for the two-class ANN model for each region, the accuracies were 95 % and 77 % for soft and hard dates, respectively. Edge detection features have a great potential in determining several surface qualities of food and agricultural products, where similar gray or color values but varying texture are found.

Study on Moving Object Detection Based on RGB Color Model

The traditional background difference method is based on gray image. Some information is lost when color image is transformed into gray image. So it is difficult to discriminate different colors with similar gray values and easily disturbed by noise and shadows. In this paper, the background difference is based on RGB color model. It is proposed to use the average value of each pixel of the color image sequences to extract the background, and then use the three-dimensional color values of the current frame and background image to compute the difference to detect the moving objects. The proposed approach is simple and easy to implement. The experimental results show that it is more sensitive to colors and has higher accuracy and robustness than the traditional background difference method. Besides, it is more resistant to shadows.

Image Segmentation Based on Poisson Equation

Image segmentation is an important part of the image processing. Currently, image segmentation methods are mainly the threshold-based segmentation method, the region-based segmentation method, the edge-based segmentation method and the Snake model based on energy function etc. This paper presents a novel image segmentation method based on the Poisson equation. The goal of the segmentation method is to divide the image into two homogeneous parts, the boundary portion and the non-boundary portion, which have similar gray values in homogeneous part. The key of the method is to build a Poisson equation with Dirichlet boundary condition. It sets a gradient threshold as the Dirichlet boundary condition of the Poisson equation, and gets a binary image by retaining the image boundary and smoothing the non-image boundary. Then simple binary segmentation will be able to get the image boundary. The experimental results show that this segmentation method can get accurate image boundaries for non-noise images and the weak noise images.

An Edge-and Region-Based Level Set Method with Shape Priors for Image Segmentation

Traditional level set methods could not deal with the segmentation of overlapped objects and the ones having the similar gray value with background,since these methods just take image data into consideration.Aiming for this problem,we propose a new edge-and region-based level set method for image segmentation.Our method firstly fuses the regional information into the edge-based level set method,and then incorporates shape priors to design a new level set method with shape priors.The comparison results show that the proposed method can segment the overlapped objects better and also achieve better performance on the objects with the similar gray value as background since it integrates multiple kinds of information together.

Segmentation of Bladder and Prostate Cancer with Biological Materials Based on Center-Focus Active Shape Model

Cancer boundary detection is an important prior task in cancer diagnosis, which will help to manage the next treatment and operation. Active shape model is proposed to solve this problem as a promising technique. But for segmentation tasks with background interference or cases such as organs with similar gray scale values, it does not always match the target contours as well as expect. This paper proposes an active shape model technique using center-focus method to robustly match the target boundaries. The algorithm has been applied on CT slices containing prostate or bladder to demonstrate the good performance of object segmentation

Segmentation of Bladder and Prostate Cancer with Biological Materials Based on Center-Focus Active Shape Model

Cancer boundary detection is an important prior task in cancer diagnosis, which will help to manage the next treatment and operation. Active shape model is proposed to solve this problem as a promising technique. But for segmentation tasks with background interference or cases such as organs with similar gray scale values, it does not always match the target contours as well as expect. This paper proposes an active shape model technique using center-focus method to robustly match the target boundaries. The algorithm has been applied on CT slices containing prostate or bladder to demonstrate the good performance of object segmentation