Contour Pixels Research Articles

Incremental component tree contour computation

A component tree is a graph representation that encodes the connected components of the upper or lower level sets of a grayscale image. Consequently, the nodes of a component tree represent binary images of the encoded connected components. There exist various algorithms that efficiently extract information and attributes of nodes of a component tree by incrementally exploiting the subset relation encoding in the tree. However, to the best of our knowledge, there is no such incremental approach to extract the contours of the nodes. In this paper, we propose an efficient incremental method to compute the contours of the nodes of a component tree by counting the edges (sides) of contour pixels. In addition, we discuss our method’s time complexity. We also experimentally show that our proposed method is faster than the standard approach based on node reconstruction.

Online detection method for magnetic suspension concentration based on machine vision

Abstract With the intelligent development of magnetic particle inspection, the quality of magnetic indications formed at cracks is closely related to the accuracy of magnetic particle inspection image analysis results. The concentration of magnetic suspension is a key process parameter affecting the quality of magnetic indication formation. Hence, this study presents an online detection method based on machine vision for measuring magnetic suspension concentration. The method initially enhances the contrast of images of the pear-shaped measuring tube containing magnetic suspension and then extracts scale lines through feature analysis and morphological processing. A method for extracting the magnetic particle sedimentation area of magnetic suspension based on a dual-threshold segmentation algorithm is proposed. The contour filtering algorithm and pixel calibration method are used to obtain the magnetic particle concentration of the non-estimation and estimation areas based on scale line extraction, ultimately forming an online accurate detection method for magnetic suspension concentration values. Experiments were conducted to validate the method against different concentrations, turbidity levels, tilting angles of the pear-shaped measuring tube, and ambient brightness. The results show that the error in magnetic suspension concentration detection based on this method is within ±3%. This has certain reference value for the stable control of magnetic suspension concentration and for enhancing the reliability of intelligent decision-making results in magnetic particle inspection.

Height estimation of sugarcane tip cutting position based on multimodal alignment and depth image fusion

Sugarcane tip cutting is essential to reducing the rate of impurities in the harvest. To achieve adaptive regulation of the tip-cutting position by a sugarcane harvester, we propose a method for estimating the height of the tip-cutting position of sugarcane. The RGB and Binocular depth cameras are aligned to process the sugarcane tip region image. This involves threshold segmentation, morphological operations, and contour detection to identify the tip-cutting position and upper boundary contours. The depth image is segmented using contour pixel information and merged to form a colour depth image of the sugarcane's tip. This image is then transformed using depth data and triangular parallax principles to determine the height of the sugarcane tip-cutting position. The proposed method was evaluated in various sugarcane plantation environments. Comparative analysis between the proposed method and manual measurements of actual cutting position heights revealed that the RMSE ranged from 1.22 cm to 1.78 cm, and R2 varied between 0.79 and 0.86. These results demonstrate the effectiveness of the proposed method in accurately extracting the height information of sugarcane tip-cutting positions, which has a specific application value for the adaptive adjustment of the tip-cutting device of the sugarcane harvester.

ИССЛЕДОВАНИЕ ПРИМЕНИМОСТИ НЕЙРОНЕЧЕТКОГО И НЕЙРОСЕТЕВОГО МОДЕЛИРОВАНИЯ ДЛЯ ВЫДЕЛЕНИЯ И РАСПОЗНАВАНИЯ КОНТУРОВ НА ИЗОБРАЖЕНИЯХ

The article describes a technique for recognizing objects in images based on neuro-fuzzy modeling of decision-making on pixel contours and neural network classification of selected object contours using features calculated by the fast Fourier transform. The proposed technique is based on solving four problems: building and using a neuro-fuzzy model for making a decision on an image pixel contour depending on the brightness difference of its eight surrounding pixels, tracking individual contours, forming a feature vector, and neural network classification of features of the contour in question. When describing the neuro-fuzzy model, the linguistic variables used, the type of fuzzy production rules, the main idea of the rule base formation algorithm, and the difference and feature of the genetic algorithm used are described. To track individual contours in a binary image, the Moore method is used, which allows obtaining an array containing sets of coordinates describing individual contours. The features of the selected contour are formed using the Fourier transform. The obtained features are invariant with respect to displacement, scaling, and rotation. The article describes the basic idea of the algorithm for constructing a training sample for a neural network classifier model. It describes the composition of the developed software package that implements the proposed method. It describes an experiment that determines the possibility of using the developed neuro-fuzzy model to detect continuous closed (without gaps) contours on a noisy image with a small PSNR value. Experimental studies on images taken from the open dataset “Kaggle - Roundabout Aerial Images for Vehicle Detection” and other sources have shown the possibility of using the developed neuro-fuzzy model for images with a “similar” distribution of the brightness gradient. It is concluded that further research is needed to determine the criterion of “similarity” of the distribution of the brightness gradient of neighboring pixels in an image for adequate application of the neuro-fuzzy model.

Detection of chronic lymphocytic leukemia using Deep Neural Eagle Perch Fuzzy Segmentation – A novel comparative approach

With a high mortality rate worldwide, Chronic Lymphocytic Leukemia (CLL) poses a serious health risk. Radiologists view the ability to detect blood tumor cells as both important and difficult. It is challenging for many of the current studies to properly examine these blood cancer cells. An intellectual system for separating healthy cells from tumor cells is suggested in this research work. This research primarily depends on the effective implementation of computed tomography (CT) samples of both benign and malignant blood samples. In this research paper, novel methodologies called public active contour pixel mapping and Deep Neural Eagle Perch Fuzzy Segmentation (DNEPFS) are proposed, which help in detecting the blood cell boundary of the unaffected cells. It is quite similar to an eagle using the claw searching algorithm. Here, the eagle uses its sharp claws to tear the prey’s skin during the search for its food. 99.64% and 99.32% accuracy are obtained during the research work for benign and malignant nodules using MATLAB R2023a. The computational complexity was found to be less, which is 7.13 s. Different segmentation techniques are used for comparative analyses. According to the research results, this proposed flow is superior to all segmentation techniques currently in use for the accurate detection of tumor cells. Based on the expert's annotation on online samples and clinical samples, the methodology has been validated to prove their effectiveness and throw light on the life of affected patients to resume normalcy – live long. The research work was tested in real-time clinical samples which deliver promising and encouraging results in leukemia detection procedures.

ENHANCED SHARP-GAN FOR HISTOPATHOLOGY IMAGE SYNTHESIS.

Histopathology image synthesis aims to address the data shortage issue in training deep learning approaches for accurate cancer detection. However, existing methods struggle to produce realistic images that have accurate nuclei boundaries and less artifacts, which limits the application in downstream tasks. To address the challenges, we propose a novel approach that enhances the quality of synthetic images by using nuclei topology and contour regularization. The proposed approach uses the skeleton map of nuclei to integrate nuclei topology and separate touching nuclei. In the loss function, we propose two new contour regularization terms that enhance the contrast between contour and non-contour pixels and increase the similarity between contour pixels. We evaluate the proposed approach on the two datasets using image quality metrics and a downstream task (nuclei segmentation). The proposed approach outperforms Sharp-GAN in all four image quality metrics on two datasets. By integrating 6k synthetic images from the proposed approach into training, a nuclei segmentation model achieves the state-of-the-art segmentation performance on TNBC dataset and its detection quality (DQ), segmentation quality (SQ), panoptic quality (PQ), and aggregated Jaccard index (AJI) is 0.855, 0.863, 0.691, and 0.683, respectively.

Foreground segmentation and location of coal and gangue under complex similar background

To improve the foreground segmentation and location accuracy of complex coal gangue images with gray histogram distribution close to the unimodal shape, a contour detection algorithm of the grayscale fluctuation matrix is proposed. The contour and non-contour pixels of coal and gangue images are investigated, and the result indicates that the gray values of the pixels around the contour exhibit the non-uniform distribution, and the gray value changes in different directions are significantly different. Accordingly, a grayscale fluctuation matrix is built by calculating the change amplitude of pixels in different directions, and multiple features are extracted from the grayscale fluctuation matrix to realize the target contour segmentation. Furthermore, the contour is optimized using the historical and future information of the contour image, thus effectively removing numerous false contours, reproducing some hidden contours and increasing segmentation accuracy. This method has high accuracy, and the maximum error rates of the pixel area and center coordinate of contour detection are 4.404% and 3.18% respectively. This study provides a feasible solution to the edge detection and segmentation of images with similar and complex backgrounds.

Fast Shape Recognition Method Using Feature Richness Based on the Walking Minimum Bounding Rectangle over an Occluded Remote Sensing Target

Remote sensing target recognition has always been an important topic of image analysis, which has significant practical value in computer vision. However, remote sensing targets may be largely occluded by obstacles due to the long acquisition distance, which greatly increases the difficulty of recognition. Shape, as an important feature of a remote sensing target, plays an important role in remote sensing target recognition. In this paper, an occluded shape recognition method based on the local contour strong feature richness (contour pixel richness, contour orientation richness, and contour distance richness) to the walking minimum bounding rectangle (MBR) is proposed for the occluded remote sensing target (FEW). The method first obtains the local contour feature richness by using the walking MBR; it is a simple constant vector, which greatly reduces the cost of feature matching and increases the speed of recognition. In addition, this paper introduces the new concept of strong feature richness and uses the new strategy of constraint reduction to reduce the complex structure of shape features, which also speeds up the recognition speed. Validation on a self-built remote sensing target shape dataset and three general shape datasets demonstrate the sophisticated performance of the proposed method. FEW in this paper has both higher recognition accuracy and extremely fast recognition speed (less than 1 ms), which lays a more powerful theoretical support for the recognition of occluded remote sensing targets.

Imaging and quantitative detection of complex defects with a flexible electromagnetic array sensor

The eddy current non-destructive testing (NDT) method has the advantages of high sensitivity and fast speed and is widely used in the inspection of key metal parts in the military and in industry. The existing scanning imaging method based on eddy current detection will cause the defect and the image to be inconsistent when the angle between the sensor and the defect is different, which will result in the inability to accurately quantify the defect based on the image. Therefore, this paper proposes a method based on a three-axis platform and a flexible eddy current array sensor to realise defect imaging and quantitative detection. First, the sensor is installed on the three-axis motion platform through a flexible substrate. It is intended that the three-axis motion platform will achieve full coverage detection of the surface of the test-piece and the overall image of the surface of the test-piece can be obtained by splicing. Then, the characteristic signal is subjected to bicubic interpolation to obtain clearer two-dimensional and three-dimensional defect images. Finally, morphological image processing is used to binarise the defect greyscale image and then the shape and size of the defect are obtained through contour pixel analysis. Experimental results show that this method can effectively overcome the impact on imaging when the sensor and the defect are at different angles and ensure the consistency of the imaging of the defect at any angle, so as to realise the imaging and quantification of any defect.

SHARP-GAN: SHARPNESS LOSS REGULARIZED GAN FOR HISTOPATHOLOGY IMAGE SYNTHESIS.

Existing deep learning-based approaches for histopathology image analysis require large annotated training sets to achieve good performance; but annotating histopathology images is slow and resource-intensive. Conditional generative adversarial networks have been applied to generate synthetic histopathology images to alleviate this issue, but current approaches fail to generate clear contours for overlapped and touching nuclei. In this study, We propose a sharpness loss regularized generative adversarial network to synthesize realistic histopathology images. The proposed network uses normalized nucleus distance map rather than the binary mask to encode nuclei contour information. The proposed sharpness loss enhances the contrast of nuclei contour pixels. The proposed method is evaluated using four image quality metrics and segmentation results on two public datasets. Both quantitative and qualitative results demonstrate that the proposed approach can generate realistic histopathology images with clear nuclei contours.

CGRNet: Contour-guided graph reasoning network for ambiguous biomedical image segmentation

In this work, we propose to address the existing problem of biomedical image segmentation that often produces results, which fail to capture the exact contours of the target and suffer from ambiguity. Most previous techniques are suboptimal because they often simply concatenate contour information to alleviate this problem, while ignoring the correlation between regions and contours. As a matter of fact, the relationship between cross-domain features is an important clue for ambiguous pixel segmentation in biomedical images. To this end, we contribute a simple yet effective framework called Contour-Guided Graph Reasoning Network (CGRNet) for more accurate segmentation against ambiguity, which is capable of capturing the semantic relations between object regions and contours through graph reasoning. Specifically, we first perform a global graph representation of the low-level and high-level features extracted by the feature extractor, where clusters of pixels with similar features are mapped to each vertex. Further, we explicitly combine contour information as the geometric prior, which can aggregate features of contour pixels to graph vertices and focus on features along the boundaries. Then, the cross-domain features propagate information through the vertices on the graph to efficiently learn and reason about the semantic relations. Finally, the learned refinement graph features are projected back to the original pixel coordinate space for the final pixel-wise segmentation task. Extensive experiments on the three publicly available Kvasir, CVC-612, and COVID19-100 datasets show the effectiveness of our CGRNet with superior performance to existing state-of-the-art methods. Our code is publicly available at: https://github.com/DLWK/CGRNet.

Drawing System with Dobot Magician Manipulator Based on Image Processing

In this paper, the robot arm Dobot Magician and the Raspberry Pi development platform were used to integrate image processing and robot-arm drawing. For this system, the Python language built into Raspberry Pi was used as the working platform, and the real-time image stream collected by the camera was used to determine the contour pixel coordinates of image objects. We then performed gray-scale processing, image binarization, and edge detection. This paper proposes an edge-point sequential arrangement method, which arranges the edge pixel coordinates of each object in an orderly manner and places them in a set. Orderly arrangement means that the pixels in the set are arranged counterclockwise to the closed curve of the object shape. This arrangement simplifies the complexity of subsequent image processing and calculation of the drawing path. The number of closed curves represents the number of strokes in the drawing of the manipulator. In order to reduce the complexity of the drawing of the manipulator, a fewer number of closed curves will be necessary. To achieve this goal, we not only propose the 8-NN (abbreviation for eight-nearest-neighbor) search, but also use to the 16-NN search and the 24-NN search methods. Drawing path points are then converted into drawing coordinates for the Dobot Magician through the Raspberry Pi platform. The structural design of the Dobot reduces the complexity of the experiment, and its attitude and positioning control can be accurately carried out through the built-in API function or the underlying communication protocol, which is more suitable for drawing applications than other fixed-point manipulators. Experimental results show that the 24-NN search method can effectively reduce the number of closed curves and the number of strokes drawn by the manipulator.

Adaptive and fast image superpixel segmentation approach

Superpixel is one of the most popular image over-segmentations with broad applications in the computer vision field to reduce their computations by replacing pixels as primitives. The main concerns of one superpixel generation algorithm are its accuracy and efficiency. One of the most important things in superpixel accuracy is to fit the image boundaries tightly with a few pixels as possible (namely minimal contour density, which is measured by the percent of superpixel contour pixels in the whole image). In this paper, we propose a new fast algorithm based on the clustering method to produce superpixels accurately with low contour density. First, we adopt the linear path from a pixel to one superpixel seed to define a regular term and propose a new distance measurement between them. In addition, we introduce the gradient and Local Binary Pattern (LBP) features and propose formulas of parameters in the proposed method adaptively. In this way, we can use the new distance measurement to group pixels as initial regions adaptively and produce the final superpixels by merging those small ones. Finally, we test the new algorithm on two public datasets and compare it with the state-of-the-art. Our method can generate superpixels with lower contour density while being competitive in accuracy and computational time.

Intelligent localization and quantitative evaluation of anterior talofibular ligament injury using magnetic resonance imaging of ankle

BackgroundThere is a high incidence of injury to the lateral ligament of the ankle in daily living and sports activities. The anterior talofibular ligament (ATFL) is the most frequent types of ankle injuries. It is of great clinical significance to achieve intelligent localization and injury evaluation of ATFL due to its vulnerability.MethodsAccording to the specific characteristics of bones in different slices, the key slice was extracted by image segmentation and characteristic analysis. Then, the talus and fibula in the key slice were segmented by distance regularized level set evolution (DRLSE), and the curvature of their contour pixels was calculated to find useful feature points including the neck of talus, the inner edge of fibula, and the outer edge of fibula. ATFL area can be located using these feature points so as to quantify its first-order gray features and second-order texture features. Support vector machine (SVM) was performed for evaluation of ATFL injury.ResultsData were collected retrospectively from 158 patients who underwent MRI, and were divided into normal (68) and tear (90) group. The positioning accuracy and Dice coefficient were used to measure the performance of ATFL localization, and the mean values are 87.7% and 77.1%, respectively, which is helpful for the following feature extraction. SVM gave a good prediction ability with accuracy of 93.8%, sensitivity of 88.9%, specificity of 100%, precision of 100%, and F1 score of 94.2% in the test set.ConclusionExperimental results indicate that the proposed method is reliable in diagnosing ATFL injury. This study may provide a potentially viable method for aided clinical diagnoses of some ligament injury.

Image Process of Rock Size Distribution Using DexiNed-Based Neural Network

In an aggregate crushing plant, the crusher performances will be affected by the variation from the incoming feed size distribution. Collecting accurate measurements of the size distribution on the conveyors can help both operators and control systems to make the right decisions in order to reduce overall power consumption and avoid undesirable operating conditions. In this work, a particle size distribution estimation method based on a DexiNed edge detection network, followed by the application of contour optimization, is proposed. The proposed framework was carried out in the four main steps. The first step, after image preprocessing, was to utilize a modified DexiNed convolutional neural network to predict the edge map of the rock image. Next, morphological transformation and watershed transformation from the OpenCV library were applied. Then, in the last step, the mass distribution was estimated from the pixel contour area. The accuracy and efficiency of the DexiNed method were demonstrated by comparing it with the ground-truth segmentation. The PSD estimation was validated with the laboratory screened rock samples.

Automated segmentation and diagnosis of pneumothorax on chest X-rays with fully convolutional multi-scale ScSE-DenseNet: a retrospective study

BackgroundPneumothorax (PTX) may cause a life-threatening medical emergency with cardio-respiratory collapse that requires immediate intervention and rapid treatment. The screening and diagnosis of pneumothorax usually rely on chest radiographs. However, the pneumothoraces in chest X-rays may be very subtle with highly variable in shape and overlapped with the ribs or clavicles, which are often difficult to identify. Our objective was to create a large chest X-ray dataset for pneumothorax with pixel-level annotation and to train an automatic segmentation and diagnosis framework to assist radiologists to identify pneumothorax accurately and timely.MethodsIn this study, an end-to-end deep learning framework is proposed for the segmentation and diagnosis of pneumothorax on chest X-rays, which incorporates a fully convolutional DenseNet (FC-DenseNet) with multi-scale module and spatial and channel squeezes and excitation (scSE) modules. To further improve the precision of boundary segmentation, we propose a spatial weighted cross-entropy loss function to penalize the target, background and contour pixels with different weights.ResultsThis retrospective study are conducted on a total of eligible 11,051 front-view chest X-ray images (5566 cases of PTX and 5485 cases of Non-PTX). The experimental results show that the proposed algorithm outperforms the five state-of-the-art segmentation algorithms in terms of mean pixel-wise accuracy (MPA) with 0.93pm 0.13 and dice similarity coefficient (DSC) with 0.92pm 0.14, and achieves competitive performance on diagnostic accuracy with 93.45% and F_1-score with 92.97%.ConclusionThis framework provides substantial improvements for the automatic segmentation and diagnosis of pneumothorax and is expected to become a clinical application tool to help radiologists to identify pneumothorax on chest X-rays.

Fast contour detection with supervised attention learning

Recent advances in deep convolutional neural networks have led to significant success in many computer vision tasks, including edge detection. However, the existing edge detectors neglected the structural relationships among pixels, especially those among contour pixels. Inspired by human perception, this work points out the importance of learning structural relationships and proposes a novel real-time attention edge detection (AED) framework. Firstly, an elaborately designed attention mask is employed to capture the structural relationships among pixels at edges. Secondly, in the decoding phase of our encoder–decoder model, a new module called dense upsampling group convolution is designed to tackle the problem of information loss due to stride downsampling. And then, the detailed structural information can be preserved even it is ever destroyed in the encoding phase. The proposed relationship learning module introduces negligible computation overhead, and as a result, the proposed AED meets the requirement of real-time execution with only 0.65M parameters. With the proposed model, an optimal dataset scale F-score of 79.5 is obtained on the BSDS500 dataset with an inference speed of 105 frames per second, which is significantly faster than existing methods with comparable accuracy. In addition, a state-of-the-art performance is achieved on the BSDS500 (81.6) and NYU Depth (77.0) datasets when using a heavier model.

Diabetic Retinopathy Feature Extraction and Classification using Adaptive Super Pixel Algorithm

Diabetic Retinopathy is an ocular manifestation of diabetes . The longer a person has diabetes, higher are the chances of having diabetic retinopathy in their visual system. Hence the objective of this research work is to propose an automated, suitable and sophisticated approach using image processing so that diabetic retinopathy can be detected at early levels easily and damage to retina can be minimized. A vital point of diabetic retinopathy that it causes detectable changes in the blood vessels of the retina. The focal blurred edges are detected so as to dismiss the false alarms. A two-level approach is used here to classify data. Firstly, optimal features are extracted from the training data and secondly, the classification is done by the use of the adaptive super pixel algorithm and then the test data is analyzed. Adaptive super pixel algorithm can adjust the weights of various features based on their discriminating ability. After the application of algorithm, the diabetic eye is detected by means of various parameters like colour, texture, spatial distance, contour, mean, standard deviation, entropy and maximum pixel points. This research can aid the doctor for easy detection of the disease as it given an accuracy of about 98.33%.

Adaptive Surface Mesh Reconstruction for Computed Tomography Images

In medical applications, it is important to reconstruct surface meshes from Computed Tomography (CT) images. Surface mesh reconstruction of biological tissues actually suffers from staircase artifacts, due to anisotropic CT data. To solve this problem, this paper proposes an adaptive surface mesh reconstruction method. We convert the contour pixels of medical image to contour points and exploit the adaptive spherical cover to produce an approximating surface based on the contour points. Due to the reconstruction quality depending on the accurate normal estimation, computing the normal vectors from the negative gradient based on 3D binary volume data instead of classical principal component analysis (PCA), and then covering contour points by adaptive spheres, linking the auxiliary points in the spheres for reconstructing adaptive triangular meshes. The presented method has been used in CT images of the first cervical vertebrae (C1), scapula, as well as the third lumbar vertebrae (L3) and the results are analyzed regarding their smoothness, accuracy and mesh quality. The results show that our method can reconstruct smooth, accurate and high-quality adaptive surface meshes.

Quality assessment for view synthesis using low-level and mid-level structural representation

View synthesis is the most important technique in multi-view and free-viewpoint videos. The whole view synthesis includes the acquisition and processing of texture and depth images, and the virtual view rendering stage. Existing quality metrics for view synthesis have limited ability for the whole synthesis process for the following reasons. First, they are dedicated to a single stage of view synthesis, overlooking the commonality of all the possible distortions introduced in the whole process. Moreover, they only extract low-level features for quality assessment, ignoring the perceptual degradation caused by the mid-level contours that are destructed by heavy distortions in texture/depth images and the imperfect view rendering, which represent the spatial distribution/connection of adjacent contour pixels. Inspired by the above facts, this paper presents a quality metric for view synthesis using both Low-level and Mid-level Structural representation (LMS), aiming to accurately evaluate the distortions in the whole view synthesis process. Specifically, the scale space is first constructed to mimic the hierarchical property of the human visual system. Then, the statistics of gradient orientation is integrated with the statistics of gradient intensity for the low-level structural representation, which is motivated by the importance of the orientation selectivity mechanism to visual perception. Further, the mid-level structure is represented using bag of words for contour description based on the sparse coding of the primary visual cortex. Then the distances of both the low-level and mid-level features between the synthesized and reference images are calculated. Finally, two distances are integrated to generate the whole quality score. Extensive experiments on two public view synthesis databases demonstrate the superiority of the proposed method to the state-of-the-arts in evaluating the quality of the whole view synthesis.