Color Image Segmentation Research Articles

Color Image Processing Under Uncertainty

Most digital images have uncertainties associated with the intensity levels of pixels and/or edges. These uncertainties can be traced back to the acquisition chain, to uneven lighting conditions used during imaging or to the noisy environment. On the other hand, intuitionistic fuzzy hypergraphs are considered a useful mathematical tool for digital image processing since they can represent digital images as complex relationships between pixels and model uncertain or imprecise knowledge explicitly. This paper presents the approach for noisy color image segmentation and edge detection based on intuitionistic fuzzy hypergraphs. First, the RGB image is transformed to the HLS space resulting in three separated components. Then each component is intuitionistically fuzzified based on entropy measure from which an intuitionistic fuzzy hypergraph is generated automatically. The generated hypergraphs will be used for denoising, segmentation, and edges detection. The first experimentations showed that the proposed approach gave good results especially in the case of dynamic threshold.

Extraction of urban buildings using color image processing and morphological operations.

In many geospatial applications, automated detection of buildings has become a key concern in recent years. Determination of building locations provides great benefits for numerous geospatial applications such as urban planning, disaster management, infrastructure planning, environmental monitoring. The study aims to present a practical technique for extracting the buildings from high-resolution satellite images using color image segmentation and binary morphological image processing. The proposed method is implemented on satellite images of 4 different selected study areas of the city of Batikent, Ankara. According to experiments conducted on the study areas, overall accuracy, sensitivity, and F1 values were computed to be on average, respectively. After applying morphological operations, the same metrics are calculated . The results show that the determination of urban buildings can be done more successfully with the suitable combination of morphological operations using rectangular structuring element. Keywords: Building Extraction; Colour Image Processing;Colour space conversion; Image Morphology; Remote Sensing

Recursive Algorithms of Maximum Entropy Thresholding on Circular Histogram

Circular histogram thresholding is a novel color image segmentation method, which makes full use of the hue component color information of the image, so that the desired target can be better separated from the background. Maximum entropy thresholding on circular histogram is one of the exist circular histogram thresholding methods. However, this method needs to search for a pair of optimal thresholds on the circular histogram of two-class thresholding in an exhaustive way, and its running time is even longer than that of the existing circular histogram thresholding based on the Otsu criteria, so the segmentation efficiency is extremely low, and the real-time application cannot be realized. In order to solve this problem, a recursive algorithm of maximum entropy thresholding on circular histogram is proposed. Moreover, the recursive algorithm is extended to the case of multiclass thresholding. A large number of experimental results show that the proposed recursive algorithms are more efficient than brute force and the existing circular histogram thresholding based on the Otsu criteria.

An improved cuckoo search algorithm for multilevel color image thresholding based on modified fuzzy entropy

Multilevel thresholding is a significant technology for image segmentation. Traditional exhaustive search method to solve the optimal multilevel thresholds is time consuming, color images contain more information are even worse. To address this issue, an improved cuckoo search algorithm (ICS) is proposed for color image segmentation in this paper, and a modified fuzzy entropy is used as its objective function. In ICS, two modifications are adopted to improve the standard cuckoo search algorithm. First, an adaptive control parameter mechanism is utilized to enhance the performance of exploration. Second, a hybrid search strategy is used to boost the local search efficiency. To fully demonstrate the superior performance of the ICS, the experiments are conducted on a series of color benchmark test images, and a total of six optimization algorithms are compared with the proposed algorithm. The experimental results show that the presented ICS algorithm outperforms all the other algorithms in term of objective function value, PSNR, FSIM, convergence speed and parametric statistical test. Compared to other algorithms, the ICS algorithm is an effective method for multilevel color image thresholding segmentation.

FAST INTERACTIVE REGIONAL PATTERN MERGING FOR GENERIC TISSUE SEGMENTATION IN HISTOPATHOLOGY IMAGES

The image segmentation of histopathological tissue images has always been a challenge due to the overlapping of tissue color distributions, the complexity of extracellular texture and the large image size. In this paper, we introduce a new region-merging algorithm, namely, the Regional Pattern Merging (RPM) for interactive color image segmentation and annotation, by efficiently retrieving and applying the user’s prior knowledge of stroke-based interaction. Low-level color/texture features of each region are used to compose a regional pattern adapted to differentiating a foreground object from the background scene. This iterative region-merging is based on a modified Region Adjacency Graph (RAG) model built from initial segmented results of the mean shift to speed up the merging process. The foreground region of interest (ROI) is segmented by the reduction of the background region and discrimination of uncertain regions. We then compare our method against state-of-the-art interactive image segmentation algorithms in both natural images and histological images. Taking into account the homogeneity of both color and texture, the resulting semi-supervised classification and interactive segmentation capture histological structures more completely than other intensity or color-based methods. Experimental results show that the merging of the RAG model runs in a linear time according to the number of graph edges, which is essentially faster than both traditional graph-based and region-based methods.

Fuzzy image clustering incorporating local and region-level information with median memberships

Image segmentation with Fuzzy C-Means (FCM) clustering algorithm is a widely researched topic. In the literature, region-level information considers more redundancy information of image and becomes a more powerful FCM technique. However, noise pixels can be over-preserved via region-level information causing FCM produce fake segmentation areas. Fuzzification of partition is also vital in FCM and Kullback–Leibler (KL) information with membership mean template is successful to reduce fuzzification. However, existence of outliers influences final segmentation result due to the sum operation during calculation of mean template. In this paper, we propose an improved FCM with adaptive local and region-level information as well as KL information with locally median membership degrees. Firstly, followed by region-level information, a new distance measure incorporating both local and region-level information is proposed where local information of the image of region-level information is acquired to smooth the over-preserved noise pixels. Secondly, we renounce the sum operation in membership mean template and propose that locally median template is more reasonable to be considered as prior probability in KL information. Thirdly, adaptive constraints for conventional FCM and local as well as region-level information are introduced. In the beginning, differences between local variances of original image and image of region-level information in exponential form are computed. Then, reciprocals of the differences and the differences themselves are considered as constraints of conventional FCM and local as well as region-level information individually. Experiments of grayscale and color image segmentation show that the proposed method FALRCM (Fuzzy Adaptive Local and Region-level information C-Means) achieves better performance in terms of fuzzy partition coefficient (VPC), fuzzy partition entropy (VPE), Segmentation Accuracy (SA), mean Intersection-over-Union (mIoU), Peak Signal-to-Noise Ratio (PSNR) and visual effects compared with several state-of-the-art FCM variants. For example, in grayscale noise image segmentation, average VPC and mIoU are up to 99.92% and 97.59% with standard deviation of 0.05% and 2.46% respectively while the method with the closest performance provides VPC and mIoU of 95.27% and 91.76% with standard deviations of 15.59% and 3.53% respectively. The proposed method FALRCM achieves better robustness of noise and lower partition fuzzification. In the end, limitations are shown in discussion and application fields as well as future studies are proposed in conclusion part.

Image segmentation with Kapur, Otsu and minimum cross entropy based multilevel thresholding aided with cuckoo search algorithm

Color image segmentation is the primary factor to provide the intended information from the input image. The straightforward method called multilevel thresholding (MLT) is used to analyse the various classes of complex images. But, when the level of threshold increases, computational difficulty increases. Hence, MLT with most promising objective functions such as Kapur, Otsu and minimum cross entropy aided with cuckoo search algorithm (CSA) is used. The efficient metaheuristic cuckoo search algorithm’s controlling parameter balances the local and global search. In this paper, the efficacy of CSA at 4,5,6 and 7 threshold levels with various fitness functions are utilized for precise image segmentation. It is seen from experimental results, the Otsu based cuckoo search algorithm outperform than Kapur and MCE. Quality metrices such as computational time, PSNR (peak signal to noise ratio) and SSIM (structural similarity index) authenticate the exploration and exploitation capability of CSA algorithm for real-world applications.

A hierarchical three-step superpixels and deep learning framework for skin lesion classification

Skin cancer is one of the most common and dangerous cancer that exists worldwide. Malignant melanoma is one of the most dangerous skin cancer types has a high mortality rate. An estimated 196,060 melanoma cases will be diagnosed in 2020 in the USA. Many computerized techniques are presented in the past to diagnose skin lesions, but they are still failing to achieve significant accuracy. To improve the existing accuracy, we proposed a hierarchical framework based on two-dimensional superpixels and deep learning. First, we enhance the contrast of original dermoscopy images by fusing local and global enhanced images. The entire enhanced images are utilized in the next step to segmentation skin lesions using three-step superpixel lesion segmentation. The segmented lesions are mapped over the whole enhanced dermoscopy images and obtained only segmented color images. Then, a deep learning model (ResNet-50) is applied to these mapped images and learned features through transfer learning. The extracted features are further optimized using an improved grasshopper optimization algorithm, which is later classified through the Naïve Bayes classifier. The proposed hierarchical method has been evaluated on three datasets (Ph2, ISBI2016, and HAM1000), consisting of three, two, and seven skin cancer classes. On these datasets, our method achieved an accuracy of 95.40%, 91.1%, and 85.50%, respectively. The results show that this method can be helpful for the classification of skin cancer with improved accuracy.

Parametric and nonparametric correlation ranking based supervised feature selection methods for skin segmentation

In supervised feature selection methods, the features that will be used for classification are selected according to their relationship with label information used for teaching classifiers. In the presented study, this relationship was measured with two different correlation coefficients, one parametric and one nonparametric. Pearson was used as the parametric correlation coefficient and Kendall was used as the nonparametric correlation coefficient. Human skin segmentation in color images has been taken as a case study to demonstrate the advantages of parametric and nonparametric correlation coefficients. In the color images, when the skin segmentation is applied based on pixels, the problem turns into a binary classification problem. For each pixel, 72 features based on color and texture are extracted. According to the order of the metrics obtained, the features are sorted by high correlated to low correlated and the strongest features are selected for classification. The classifier is trained and tested by extracted features with and without feature selection methods obtained from the Pratheepan database. When the eliminated features are added again, it is seen that either its contribution is negligibly low or it disrupts the classification.

Color Image Segmentation and Detection of Barrels Using Gaussian Mixture Model

This paper is based on segmentation and detection of the color barrels. The main goal is to identify the barrels, there may be one or more barrels in the test and the training set in the given RGB images after managing segmentation color. Besides, Gaussian Mixture Model, Bayesian Model, linear regression method and other techniques are used to obtain more precise results of training images as well as prediction. We identify different color barrels such as red, blue, green and the barrel mean, variance and distance is approximately estimated by importing functions in this class by using linear regression model. GMM that is modeled as pixels and color classes, for the categorization of pixels by the class of color, and then we can differentiate the area of pixels of the barrel from non-barrel. The distance is estimated by two purposes interior One to detect the barrel and other to find the bounding box information counting four vertexes, height and width of the barrel.

Color image segmentation using Kapur, Otsu and Minimum Cross Entropy functions based on Exchange Market Algorithm

Color image segmentation is the primary step to elicit detailed information of the image with RGB color space. The key to acquire the region of interest is the straightforward, instinctual technique called thresholding. Bi-level thresholding can be assessed concretely for simple images, whereas revealing various classes from a complex image is achieved only through multiple threshold values. Hence, multilevel thresholding with the most propitious maximizing objective functions such as Kapur, Otsu and Minimum Cross Entropy (MCE) to extract the precise threshold are employed in this paper. As the segmentation level increases, execution of non-parametric objective functions leads to exponential increase in computational time. Several researches to enhance the speed of objective functions along with various metaheuristics such as Krill Herd Algorithm (KHA), Teaching-Learning Based Optimization (TLBO) and Cuckoo Search Algorithm (CSA) are performed. The need to unlock the computational complexity drifted our view towards the most powerful, robust, recently introduced metaheuristic Exchange Market Algorithm (EMA). EMA involves exchange of shares among investors to gain profit. The strategic approach of share members in stable and unstable modes of the market results in precise exploration and exploitation. Furthermore, low fitness shareholders gaining experience from their high and moderate counterparts, stands out as an exceptional step to reduce the time consumption. This concept is utilized in our paper for the first time to obtain the exact details from the image without any delay. Empirical outcome of the results indicates that exceptional image segmentation is achieved by EMA in less time compared to extensive search techniques such as KHA, TLBO and CSA. Quantitative and qualitative validation furnished by metrics such as Structural Similarity Index (SSIM), computational time, Peak Signal to Noise Ratio (PSNR) and statistical Wilcoxon test affirm that Otsu, Kapur and MCE based EMA outperforms the existing techniques to analyze the real-world images.

Characterization of combustion process and emissions in a natural gas/diesel dual-fuel compression-ignition engine

Dual-fuel (DF) combustion using natural gas and diesel has received considerable attention in the on- and off-road freight transportation sectors owing to its potential use in achieving better fuel economy and cleaner combustion. To determine the effect of the natural gas/diesel mixture quality on the combustion process and pollutant emission, high-speed flame visualization was used to investigate the phenomena of natural gas (NG)/diesel DF combustion in a 1.0 L optically-accessible single-cylinder engine. The diesel injection timing and natural gas substitution ratio (NGSR) were varied to implement diverse in-cylinder blending conditions under constant fuel energy input. A novel flame regime separation method based on color image segmentation in a hue-saturation-value (HSV) color space was used to quantitatively compare the spatial distributions of premixed and diffusion flame regimes. Because NG has a lower carbon content and higher auto-ignition resistance compared with diesel, the natural luminosity images for larger NGSRs revealed a significant reduction in the diffusion flame regime accompanied by retarded flame development. An earlier injection of the liquid diesel shifted the location of the early flame growth toward the piston bowl wall and created a rapid influx of propagating flame, while effectively suppressing the formation of intense soot radiation through longer ignition delay. These observations were verified by the exhaust emissions measured using the full-metal version of the engine and the same fuel supply parameters, specifically regarding the behavior of the smoke and nitrogen oxide emissions.

Colour image segmentation based on a convex K‐means approach

Image segmentation is a fundamental and challenging task in image processing and computer vision. The colour image segmentation is attracting more attention as the colour image provides more information than the grey image. A variational model based on a convex K-means approach to segment colour images is proposed. The proposed variational method uses a combination of l1 and l2 regularizers to maintain edge information of objects in images while overcoming the staircase effect. Meanwhile, our one-stage strategy is an improved version based on the smoothing and thresholding strategy, which contributes to improving the accuracy of segmentation. The proposed method performs the following steps. First, the colour set which can be determined by human or the K-means method is specified. Second, a variational model to obtain the most appropriate colour for each pixel from the colour set via convex relaxation and lifting is used. The Chambolle–Pock algorithm and simplex projection are applied to solve the variational model effectively. Experimental results and comparison analysis demonstrate the effectiveness and robustness of the method.

A multiverse optimization based colour image segmentation using variational mode decomposition

Multilevel thresholding using the histogram is the most popular and accepted technique of image segmentation. The computational time of multilevel thresholding rises exponentially as the number of the thresholds increases. Histogram suffers from irregularities and sharp details which leads to stagnation. In this article, a newly developed multiverse optimization is combined with variational mode decomposition to overcome this problem. The histogram of an input image divided into several band-limit modes using VMD then reconstruct histogram using meaningful modes. The reconstructed histogram is free from the high-frequency fluctuation which causes local optima. The proposed method utilized two entropy function to develop image segmentation by determining the optimal threshold. The result of the proposed algorithm is analyzed with other evolutionary algorithms such as artificial bee colony, sine cosine algorithm, and salp swarm algorithm. Comparison is made based on comparative parameters such as peak signal to noise ratio, structural similarity index, feature similarity index, uniformity, normalized absolute error, quality index based on local variance, computational time, and mean square error. The test results validate that the proposed algorithm presents more reliable results than other existing techniques.

Multiobjective fuzzy clustering with multiple spatial information for Noisy color image segmentation

Clustering method is a widely used and effective technique in color image segmentation. In general, traditional clustering-based image segmentation algorithms consider only one objective function and the segmentation performance is easily influenced by the noise in the image. Therefore, utilizing many clustering criteria and the neighborhood statistic information of pixels are more effective to improve the segmentation performance. In this paper, we put forward a multiobjective fuzzy clustering algorithm with multiple spatial information (MFCMSI) for noisy color image segmentation. Firstly, two conflicting fitness functions including the local spatial information with a circular neighborhood window and the non-local spatial information with circular search and similarity windows are designed to improve the noise-resistant ability and segmentation performance. In order to optimize these two fitness functions, the variable-length and cluster-center-based encoding strategy and some efficient evolutionary operations are utilized in each generation. Finally, a cluster validity index with multiple spatial information is constructed to select the best solution from the final non-dominated solution set of the last generation. Aiming to improve the robustness of MFCMSI, the ensemble strategy is introduced into MFCMSI and an ensemble version of MFCMSI is presented. Experimental results show that MFCMSI and its ensemble version behave well in evolving the number of segments and obtaining the satisfactory segmentation performance.

A Type-2 Fuzzy Clustering and Quantum Optimization Approach for Crops Image Segmentation

Automatic detection of crop yield ripeness is a tedious task because of the presence of various intensities of color in crops. One of the solutions to this problem is the monitoring of those crops by performing segmentation operations. This operation can help to distinguish the ripe and non-ripe regions among the crop images. For this purpose, this study presents a new hybrid crop image segmentation method utilizing type-2 fuzzy set (T2FS), K-means clustering algorithm, and modified quantum optimization algorithm (MQOA). The proposed method fully utilizes the indispensable qualities of these three techniques by (a) using T2FS to represent each color component of crop images in terms of secondary memberships, (b) applying K-means clustering algorithm to extract the similar features from the set of type-2 entropy values obtained from the secondary memberships, and (c) exploiting MQOA to optimize the distance function used in K-means clustering algorithm to obtain the optimal clusters. The performance of the proposed method is assessed based on the experiments carried out on color images of cherry tomatoes. Evidence of experimental results suggests that the proposed method produces extremely effective segmented images relative to those well-known color image segmentation methods available in the literature of pattern recognition and computer vision domains.

Adaptive scale weighted fuzzy C-Means clustering for the segmentation of purple soil color image

The segmentation and extraction of the purple soil region from purple soil color image can effectively avoid the influence of background on recognition of soil types. A scale weighted fuzzy c-means clustering algorithm(SWFCM) is proposed for effective segmentation of purple soil color image. The main work is to establish the maximum difference optimization model with the mean of Gaussian distance between each pixel and each peak of the image histogram, and optimize the clustering number and the initial clustering centers. Then, the compactness of each class is defined to weight the Euclidean distance between the pixel and the clustering center and improve the optimization model of FCM for raising its clustering performance. Aiming at the problem of removing scattered small soil blocks in the background and filling holes in the purple soil region, the algorithm of extracting the boundary of the purple soil region and the algorithm of filling the purple soil region are proposed. Finally, the normal and robust experiments are carried out on the normal sample set and robust sample set. And the performances of relative algorithms are compared, which involves the previously released FCM algorithms and some methods for the segmentation of purple soil color image and our proposed algorithm. Experimental results show that performance of SWFCM is better and it can provide a high reference for adaptive segmentation of purple soil color images. Especially for robust experiment images, its average segmentation accuracy is improved by 6 . 64% ∼ 8 . 25 % compared with other purple soil segmentation algorithms.

Multilevel thresholding using a modified ant lion optimizer with opposition-based learning for color image segmentation

Multilevel thresholding has important research value in image segmentation and can effectively solve region analysis problems of complex images. In this paper, Otsu and Kapur's entropy are adopted among thresholding segmentation methods. They are used as the objective functions. When the number of threshold increases, the time complexity increases exponentially. In order to overcome this drawback, a modified ant lion optimizer algorithm based on opposition-based learning (MALO) is proposed to determine the optimum threshold values by the maximization of the objective functions. By introducing the opposition-based learning strategy, the search accuracy and convergence performance are increased. In addition to IEEE CEC 2017 benchmark functions validation, 11 state-of-the-art algorithms are selected for comparison. A series of experiments are conducted to evaluate the segmentation performance of the algorithm. The evaluation metrics include: fitness value, peak signal-to-noise ratio, structural similarity index, feature similarity index, and computational time. The experimental data are analyzed and discussed in details. The experimental results significantly demonstrate that the proposed method is superior over others, which can be considered as a powerful and efficient thresholding technique.

A hybrid equilibrium optimizer algorithm for multi-level image segmentation.

Threshlod image segmentation is a classic method of color image segmentation. In this paper, we propose a hybrid equilibrium optimizer algorithm for multi-level image segmentation. When multi-level threshold method calculates the neighborhood mean and median of color image, it takes huge challenge to find the optimal threshold. We use the proposed method to optimize the multi-level threshold method and get the optimal threshold from the color image. In order to test the performance of the proposed method, we select the CEC2015 dataset as the benchmark function. The result shows that the proposed method improves the optimization ability of the original algorithm. And then, the classic images and wood fiber images are taken as experimental objects to analyze the segmentation result. The experimental results show that the proposed method has a good performance in Uniformity measure, Peak Signal-to-Noise Ratio and Feature Similarity Index and CPU time.

Ensemble Method of Deep Learning, Color Segmentation, and Image Transformation to Track, Localize, and Count Cotton Bolls Using a Moving Camera in Real-Time

HighlightsAn ensemble method using color segmentation, deep learning, and image transformation was developed.Experiments were conducted to compare the method with other state-of-the-art tracking algorithms.The optimized ensemble method to track bolls achieved 94.4% accuracy using weakly trained tiny YOLOv2 models.The method achieved 7.6 frames per second and outperformed five other tracking methods.Abstract. In robotic applications, good perception can be computationally costly and create undesirable latency before a control decision is initiated. Most of the methods available for object detection deep learning are either fast with low accuracy or slow with high accuracy. Fast and accurate methods are necessary to track and localize objects such as cotton bolls that may be visible or occluded by each other or not well illuminated. In this study, an ensemble of a deep learning method and other image processing techniques was used to detect cotton bolls in-field on defoliated plants. In each image, a trained deep learning method, the YOLOv2 model, was used to detect open cotton bolls, and color segmentation was applied to confirm if the bolls detected by the YOLOv2 model were actually white to avoid false positives. Boll tracking was performed by following the spatial movement of good features on the edges of the bolls using the Lucas-Kanade algorithm. An image transformation algorithm was applied to the next image in case the previously detected boll was lost to retrieve the information of the missing boll. Each tracked and localized boll was stored and counted to give the total number of bolls detected. In this study, detection accuracy was sacrificed for image processing speed by using the YOLOv2 model. Detection accuracy was improved by using an ensemble method that combined image color segmentation, optical flow, and image transformation. This method was compared to eight other open-source methods implemented in OpenCV. The ensemble method detected and counted bolls at a speed of 7.6 fps with an accuracy of 94.4% using the Jetson TX2 embedded system to process 1K resolution images, outperforming the other OpenCV methods in various measurements. Keywords: Boll counting, Cotton, Cotton harvesting, DarkFlow, Darknet, Deep learning, Machine vision, YOLOv2.