Image Segmentation Techniques Research Articles

Research progress on medical ultrasound image segmentation algorithms

Medical ultrasound imaging is an integral part of preoperative diagnosis, lesion screening and ultrasound-guided interventional surgeries. Image segmentation techniques can enhance the identification of lesions and separate them from complex backgrounds, aiding physicians in both quantitative and qualitative analyses. Ultrasound image segmentation algorithms are primarily categorized into two types: traditional non-semantic segmentation and deep learning-based semantic segmentation, each with distinct advantages and drawbacks. This paper delves into these segmentation principles, elucidating their relevance in the realm of ultrasound image segmentation, and offers an overview of current research trends. Our goal is to provide guidance for physicians and researchers in selecting the most suitable segmentation algorithm that tailors to their specific requirements.

Detection of Kidney Stone and Estimation of its Size using Image Segmentation Techniques

The study discusses the development of a reliable strategy for detecting and segmenting kidney stones from CT scans. Kidney stones are solid materials that can form in the kidneys due to excessive amounts of specific minerals in the urine. Severe back, side, lower abdominal, or groin discomfort, or blood in urine can indicate kidney stone presence. The current method for detection using CT scans is widely used, but it requires more precise and efficient technology due to the laborious and time-consuming manual process. The objective of this report is to propose a robust approach for kidney stone detection and segmentation, employing threshold segmentation and the watershed algorithm along with pre-and post-image processing for noise reduction. Additionally, the method estimates the size of the stone, aiding in selecting the appropriate medical procedure for stone removal. The entire process is implemented on a dataset comprising numerous kidney stone images to test and validate its accuracy and reproducibility. The system's image quality enhancement is evaluated by calculating several parameters such as SSIM, PSNR, FSIM, NCC, and NAE on the output images. To further validate the suggested method, a comparison is made with two other existing algorithms based on the mentioned evaluation criteria. The proposed strategy promises to improve the accuracy and reliability of kidney stone detection and segmentation, providing a more efficient and effective solution for medical diagnosis and treatment

Hybrid Convolutional Neural Network with Whale Optimization Algorithm (HCNNWO) Based Plant Leaf Diseases Detection

Plant diseases appear to be posing a serious danger to the production and availability of food globally. The main factor affecting the quality and productivity of agricultural products is the health of the plants. In this paper, we describe a modified plant disease detection using deep convolutional neural networks in real time. By employing image processing techniques to enlarge the plant illness photos, the plant disease sets of data were initially produced. To recognise plant illnesses, a system called Convolutional Neural Network combined with Wolf Optimisation algorithm (CNN-WO) was used. Finally, the Whale Optimization algorithm (WO) is used to maximise and optimizes getting input. And it is given to CNN's learning rate for classification process. This paper presents an image segmentation and classification technique to automatically identify plant leaf diseases. The suggested strategy increased accuracy, sensitivity, precision, F1 measure, and specificity of plant disease detection. According to this study, HCNNWO real detectors have improved, which would require deep learning. It would be an effective method for determining plant illnesses and other diseases within plants. According to the evaluation report, the suggested method offers good reliability. To evaluate how well the suggested algorithm performs in comparison to cutting-edge techniques such as SVM, BPNN and CNN, experiments are conducted on datasets that are openly accessible

A Novel Image Segmentation Technique with N-L Means Filter Using Region Merging Strategy

Image segmentation is a branch of digital image processing which emphases on splitting an image into different fragments according to their features. This work proposes a novel image segmentation technique using fast non local (N-L) means filter with region merging approach. Here, a random noise is added to the original image and discrete wavelet transform (DWT) is applied to separate the low and high frequency bands of the digital image. The low frequency component is subjected to a fast N-L means filter and the filtered image is segmented by graph-based segmentation technique. Here, these segmented image regions are merged to reduce the over segmentation patterns. The detailed coefficients of the image component are treated with soft threshold filter to reduce the residual noise. Finally merged image is combined with preserved edge features at wavelet projection resulting in better segmented image. The results are compared with some of the best methods available in literature and are found to be significantly better. Keywords: Discrete wavelet transform (DWT), fast NL-means filter, graph-based segmentation, boundary based hierarchical region merging, soft-thresholding, peak signal to noise ratio (PSNR).

Novel modified kernel fuzzy c-means algorithm used for cotton leaf spot detection

Image segmentation is a significant and difficult subject that is a prerequisite for both basic image analysis and sophisticated picture interpretation. In image analysis, picture segmentation is crucial. Several different applications, including those related to medicine, facial identification, Cotton disease diagnosis, and map object detection, benefit from image segmentation. In order to segment images, the clustering approach is used. The two types of clustering algorithms are Crisp and Fuzzy. Crisp clustering is superior to fuzzy clustering. Fuzzy clustering uses the well-known FCM approach to enhance the results of picture segmentation. KFCM technique for image segmentation can be utilized to overcome FCM’s shortcomings in noisy and nonlinear separable images. In the KFCM approach, the Gaussian kernel function transforms high-dimensional, nonlinearly separable data into linearly separable data before applying FCM to the data. KFCM is enhancing noisy picture segmentation results. KFCM increases the accuracy rate but ignores neighboring pixels. The Modified Kernel Fuzzy C-Means approach is employed to get over this problem. The NMKFCM approach enhances picture segmentation results by including neighboring pixel information into the objective function. This suggested technique is used to find “blackarm” spots on cotton leaves. A fungal leaf disease called “blackarm” leaf spot results in brown leaves with purple borders. The bacterium can harm cotton plants, causing angular leaf blotches that range in color from red to brown.

Machine learning and deep learning for brain tumor MRI image segmentation.

Brain tumors are often fatal. Therefore, accurate brain tumor image segmentation is critical for the diagnosis, treatment, and monitoring of patients with these tumors. Magnetic resonance imaging (MRI) is a commonly used imaging technique for capturing brain images. Both machine learning and deep learning techniques are popular in analyzing MRI images. This article reviews some commonly used machine learning and deep learning techniques for brain tumor MRI image segmentation. The limitations and advantages of the reviewed machine learning and deep learning methods are discussed. Even though each of these methods has a well-established status in their individual domains, the combination of two or more techniques is currently an emerging trend.

SEMANTIC ENRICHMENT OF 3D POINT CLOUDS USING 2D IMAGE SEGMENTATION

Abstract. 3D point cloud segmentation is computationally intensive due to the lack of inherent structural information and the unstructured nature of the point cloud data, which hinders the identification and connection of neighboring points. Understanding the structure of the point cloud data plays a crucial role in obtaining a meaningful and accurate representation of the underlying 3D environment. In this paper, we propose an algorithm that builds on existing state-of-the-art techniques of 2D image segmentation and point cloud registration to enrich point clouds with semantic information. DeepLab2 with ResNet50 as backbone architecture trained on the COCO dataset is used for indoor scene semantic segmentation into several classes like wall, floor, ceiling, doors, and windows. Semantic information from 2D images is propagated along with other input data, i.e., RGB images, depth images, and sensor information to generate 3D point clouds with semantic information. Iterative Closest Point (ICP) algorithm is used for the pair-wise registration of consecutive point clouds and finally, optimization is applied using the pose graph optimization on the whole set of point clouds to generate the combined point cloud of the whole scene. 3D point cloud of the whole scene contains pseudo-color information which denotes the semantic class to which each point belongs. The proposed methodology use an off-the-shelf 2D semantic segmentation deep learning model to semantically segment 3D point clouds collected using handheld mobile LiDAR sensor. We demonstrate a comparison of the accuracy achieved compared to a manually segmented point cloud on an in-house dataset as well as a 2D3DS benchmark dataset.

Discrete Wavelet Analysis: A Mighty Approach for Image Segmentation

This paper explores the application of Discrete Wavelet Analysis, a mathematical and signal processing technique, in the context of image segmentation, which provide a pixel-level or region-level decomposition of the image, enabling the extraction of relevant information for subsequent analysis and interpretation. Introducing the basic image segmentation techniques and the DWA, this paper discovers that DWA has found widespread application in fields such as signal processing, image analysis, and data compression. Compared with Fourier Transform, DWA is more suitable for image segmentation, having unique advantages and characteristics. Among the procedures of image segmentation, the most important point is feature selection, which determine the criteria for distinguishing different regions within the image. Despite DWA has many advantages, this technology also owns many challenges and limitations, which may be solved by lasting academic research to refine and extend Discrete Wavelet Analysis methodologies for image segmentation. In short, this research highlights the promise of Discrete Wavelet Analysis, emphasizing the use of high- quality image processing.

A Novel Fuzzy-Based Remote Sensing Image Segmentation Method.

Image segmentation is a well-known image processing task that consists of partitioning an image into homogeneous areas. It is applied to remotely sensed imagery for many problems such as land use classification and landscape changes. Recently, several hybrid remote sensing image segmentation techniques have been proposed that include metaheuristic approaches in order to increase the segmentation accuracy; however, the critical point of these approaches is the high computational complexity, which affects time and memory consumption. In order to overcome this criticality, we propose a fuzzy-based image segmentation framework implemented in a GIS-based platform for remotely sensed images; furthermore, the proposed model allows us to evaluate the reliability of the segmentation. The Fast Generalized Fuzzy c-means algorithm is implemented to segment images in order to detect local spatial relations between pixels and the Triple Center Relation validity index is used to find the optimal number of clusters. The framework elaborates the composite index to be analyzed starting by multiband remotely sensed images. For each cluster, a segmented image is obtained in which the pixel value represents, transformed into gray levels, the graph belonging to the cluster. A final thematic map is built in which the pixels are classified based on the assignment to the cluster to which they belong with the highest membership degree. In addition, the reliability of the classification is estimated by associating each class with the average of the membership degrees of the pixels assigned to it. The method was tested in the study area consisting of the south-western districts of the city of Naples (Italy) for the segmentation of composite indices maps determined by multiband remote sensing images. The segmentation results are consistent with the segmentations of the study area by morphological and urban characteristics, carried out by domain experts. The high computational speed of the proposed image segmentation method allows it to be applied to massive high-resolution remote sensing images.

TinyML Olive Fruit Variety Classification by Means of Convolutional Neural Networks on IoT Edge Devices

Machine learning (ML) within the edge internet of things (IoT) is instrumental in making significant shifts in various industrial domains, including smart farming. To increase the efficiency of farming operations and ensure ML accessibility for both small and large-scale farming, the need for a low-cost ML-enabled framework is more pressing. In this paper, we present an end-to-end solution that utilizes tiny ML (TinyML) for the low-cost adoption of ML in classification tasks with a focus on the post-harvest process of olive fruits. We performed dataset collection to build a dataset that consists of several varieties of olive fruits, with the aim of automating the classification and sorting of these fruits. We employed simple image segmentation techniques by means of morphological segmentation to create a dataset that consists of more than 16,500 individually labeled fruits. Then, a convolutional neural network (CNN) was trained on this dataset to classify the quality and category of the fruits, thereby enhancing the efficiency of the olive post-harvesting process. The goal of this study is to show the feasibility of compressing ML models into low-cost edge devices with computationally constrained settings for tasks like olive fruit classification. The trained CNN was efficiently compressed to fit into a low-cost edge controller, maintaining a small model size suitable for edge computing. The performance of this CNN model on the edge device, focusing on metrics like inference time and memory requirements, demonstrated its feasibility with an accuracy of classification of more than 97.0% and minimal edge inference delays ranging from 6 to 55 inferences per second. In summary, the results of this study present a framework that is feasible and efficient for compressing CNN models on edge devices, which can be utilized and expanded in many agricultural applications and also show the practical insights for implementing the used CNN architectures into edge IoT devices and show the trade-offs for employing them using TinyML.

Digital Cake Analysis: A Novel Coal Filter Cake Examination Technique Using Microcomputed Tomography

Filtration is crucial for separating solids and liquids in various industries. Understanding slurry properties and filter cake structures is essential for optimising filtration performance. Conventional methods focus on interpreting filtration data to improve the understanding of filtration mechanisms. However, examining fragile filter cakes is challenging, and current techniques often alter their structure. Conventional methods only provide an average representation of cake porosity, neglecting variations in porosity and pore distribution across the cake thickness. This study introduces the Digital cake analysis, a non-destructive technique for evaluating filter cake structure. Filtration experiments using a custom-built unit were conducted on coal slurries to obtain filter cake samples. X-ray-microcomputed tomography (µCT) imaging was utilized for cake analysis. Image enhancement techniques were employed to improve the quality of the µCT images. The enhanced images were segmented into three phases (resolved pore, subresolution pore, and solid phases) for quantitative analysis. This segmentation technique allocated partial pore volume to voxels in the subresolution phase based on their intermediate grey-scale intensity, enabling more accurate porosity calculations. Unlike conventional methods, this approach computed porosity values for resolved (100% void) and subresolution (partially void) pores. This image segmentation technique facilitated accurate computations of porosity, pore size distribution, and pore properties, significantly advancing the understanding of cake structures. Digital cake analysis produced porosity measurements similar to the experimental results.

Combining contour-based and region-based in image segmentation

Background This paper presents an optimized clustering approach applied to image segmentation. Accurate image segmentation impacts many fields like medical, machine vision, object detection. Applications involve tumor detection, face detection and recognition, and video surveillance. Methods The developed approach is based on obtaining an optimum number of clusters and regions of an image. We combined Region-based and contour-based approaches. Initial rough regions are obtained using edge detection. We have used Gabor wavelets for texture classification and spatial resolutions. Color frequencies are also used to determine the number of clusters of the Fuzzy c-means (FCM) algorithm which gave an optimum number of clusters or regions. Results We have compared our approach with other similar wavelet and clustering techniques. Our algorithm gave better values for segmentation metrics like SNR, PSNR, and MCC. Conclusions Optimizing the number of clusters or regions has a significant effect on the performance of the image segmentation techniques. This will result in better detection and localization of the segmentation-based application.

Exploring a novel HE image segmentation technique for glioblastoma: A hybrid slime mould and differential evolution approach

Glioblastoma is a primary brain tumor with high incidence and mortality rates, posing a significant threat to human health. It is crucial to provide necessary diagnostic assistance for its management. Among them, Multi-threshold Image Segmentation (MIS) is considered the most efficient and intuitive method in image processing. In recent years, many scholars have combined different metaheuristic algorithms with MIS to improve the quality of Image Segmentation (IS). Slime Mould Algorithm (SMA) is a metaheuristic approach inspired by the foraging behavior of slime mould populations in nature. In this investigation, we introduce a hybridized variant named BDSMA, aimed at overcoming the inherent limitations of the original algorithm. These limitations encompass inadequate exploitation capacity and a tendency to converge prematurely towards local optima when dealing with complex multidimensional problems. To bolster the algorithm’s optimization prowess, we integrate the original algorithm with a robust exploitative operator called Differential Evolution (DE). Additionally, we introduce a strategy for handling solutions that surpass boundaries. The incorporation of an advanced cooperative mixing model accelerates the convergence of BDSMA, refining its precision and preventing it from becoming trapped in local optima. To substantiate the effectiveness of our proposed approach, we conduct a comprehensive series of comparative experiments involving 30 benchmark functions. The results of these experiments demonstrate the superiority of our method in terms of both convergence speed and precision. Moreover, within this study, we propose a MIS technique. This technique is subsequently employed to conduct experiments on IS at both low and high threshold levels. The effectiveness of the BDSMA-based MIS technique is further showcased through its successful application to the medical image of brain glioblastoma. The evaluation of these experimental outcomes, utilizing image quality metrics, conclusively underscores the exceptional efficacy of the algorithm we have put forth.

PU-NET DEEP LEARNING ARCHITECTURE FOR GLIOMAS BRAIN TUMOUR SEGMENTATION IN MAGNETIC RESONANCE IMAGES

Automatic medical image segmentation is one of the main tasks for many organs and pathology structure delineation. It is also a crucial technique in the posterior clinical examination of brain tumours, like applying radiotherapy or tumour restrictions. Various image segmentation techniques have been proposed and applied to different image types of images. Recently, it has been shown that the deep learning approach accurately segments images, and its implementation is usually straightforward. In this paper, we proposed a novel approach, called PU-net, for automatic brain tumour segmentation in multi-modal magnetic resonance images (MRI) based on deep learning. We introduced an input processing block to a customised fully convolutional network derived from the U-Net network to handle the multi-modal inputs. We performed experiments over the Brats brain tumour dataset collected in 2018 and achieved dice scores of 0.905,0.827, and 0.803 for the whole tumour (WT), tumour core (TC), and enhancing tumour (ET) classes, respectively. This study also provides promising results compared to the traditional machine learning methods, such as support vector machines (SVM), random forest (RF) and other deep learning methods used in this context.

RETRACTED: Deep learning network (DL-Net) based classification and segmentation of multi-class retinal diseases using OCT scans

The most frequent cause of blindness in industrialized nations is age-related macular degeneration (AMD), particularly in persons over 60. In recent years in this field, three key factors have led to an increase in the workload of experts and the healthcare system: 1) population aging worldwide prevalence, 2) AMD's chronic nature, and 3) more widespread usage of the retinal optical coherence tomography (OCT) imaging technology. The establishment of fully automated diagnosis systems has recently been made possible by recent developments in the deep learning field. The main objective of the research is to develop a new deep learning-based model using OCT images. The performance for classification and segmentation of retinal diseases is improved by the proposed model through effectively capturing inter-scale features and combining these features using convolutional blocks. Then, we classify volumetric OCT images via a new deep learning-based model. For multi-class classification and segmentation of ocular diseases, the OCT scans of the human eye are used in this research to demonstrate a Deep Learning Network (DL-Net) model by automatically recognizing normal, diabetic macular edema (DME), choroidal neovascularization (CNV), AMD, and drusen images. This research introduces the Modified ResNet-50 approach and Image Processing for multiple OCT image classification. This research presents an efficient diagnostic technique for image segmentation based on a Bi-LSTM-based deep recurrent convolutional neural network (DRCNN). The modified SqueezeNet model analyzes the volumetric segmentation of OCT images. The volumetric publicly available datasets are used to analyze the proposed model. The proposed multi-scale structure model outperforms several well-known OCT classification and segmentation frameworks. 99.76% accuracy is achieved by the proposed model respectively. In healthcare settings, the proposed methodology can be used as a diagnostic instrument to help medical experts make better-diagnosing decisions, as shown by the encouraging proposed architecture's qualitative evaluations and quantitative results by generating a confusion matrix.

Aquila-particle swarm based cooperative search optimizer with superpixel techniques for epithelial layer segmentation

The segmentation of epithelial layers from oral histopathology images plays a crucial role for early detection of oral cancer disease. As a result, more accurate segmentation of this layer is of the utmost importance for a Computer Aided Diagnosis (CAD) system. Therefore, this paper presents a superpixel image-based clustering technique using an improved Nature-Inspired Optimization Algorithm (NIOA), called the Cooperative Search (CS) algorithm. Here, superpixel image is utilized to construct a faster and noise-resistant oral histopathology image clustering method. The designed CS, on the other hand, is based on two popular NIOAs, the Aquila Optimizer (AO) and the Particle Swarm Optimizer (PSO), has better exploration-exploitation abilities. This CS has been used in the image clustering field to circumvent the problem of local optima trapping. On the other hand, a comparative investigation of three popular superpixel strategies with the proposed CS optimizer has been conducted to determine the optimal superpixel strategy for epithelial layer segmentation from clean as well as noisy oral histopathology images. Finally, the optimal superpixel strategy with CS has been tested with other state-of-the-art image segmentation techniques in the epithelial layer segmentation domain. The results obtained by the proposed segmentation technique are 98.82%, 96.70%, 97.60%, and 95.25%, in terms of average Accuracy, MCC, Dice, and Jaccard, respectively, which all are better than the cutting-edge segmentation methods. The proposed segmentation model is also evaluated over the leaf segmentation model and achieved better results visually and numerically. The optimization efficiency of the CS has also been measured over CEC2019 mathematical benchmark test functions and produced competitive results compared to other tested NIOAs.

Combining contour-based and region-based in image segmentation

Background: This paper presents an optimized clustering approach applied to image segmentation. Accurate image segmentation impacts many fields like medical, machine vision, object detection. Applications involve tumor detection, face detection and recognition, and video surveillance. Methods: The developed approach is based on obtaining an optimum number of clusters and regions of an image. We combined Region-based and contour-based approaches. Initial rough regions are obtained using edge detection. We have used Gabor wavelets for texture classification and spatial resolutions. Color frequencies are also used to determine the number of clusters of the Fuzzy c-means (FCM) algorithm which gave an optimum number of clusters or regions. Results: We have compared our approach with other similar wavelet and clustering techniques. Our algorithm gave better values for segmentation metrics like SNR, PSNR, and MCC. Conclusions: Optimizing the number of clusters or regions has a significant effect on the performance of the image segmentation techniques. This will result in better detection and localization of the segmentation-based application.

Hybrid conservative central/WENO finite difference scheme for two-dimensional detonation problems

In the present work, a hybrid finite difference scheme is proposed and tested for the study of two-dimensional detonation problems. The hybrid scheme consists of a fifth-order weighted essentially nonoscillatory (WENO) method for discontinuous regions and a fourth or sixth-order robust conservative central finite difference scheme for the remaining domain. An arbitrary shock sensor is used to identify the discontinuities. The proposed shock sensor is based on the absolute value of the density gradient with an arbitrary threshold computed at each time step using a fast image segmentation technique. The sensor is complemented with a Ducros sensor to avoid the selection of vortices as discontinuities. The hybrid scheme is tested for several benchmark examples, including a two-dimensional transverse detonation wave, detonation wave diffraction, and a multiple obstacles test case. The obtained results show that the proposed sensor detects discontinuities adequately and the hybrid schemes give the same results as the fifth-order WENO in faster computational time.

Stereo vision-based measurement of wave evolution around square column in laboratory

The spatial-temporal measurement of complex wave evolution is significant in studying wave-structure interactions. Current methods, such as that using wave probes, have shown limitations in measuring the wave evolution around structures in laboratories. In this study, an improved stereo imaging method is proposed for measuring the wave evolution around a fixed structure. Regular wave tests were conducted on a fixed surface-piercing square column in a wave flume to validate the reliability and accuracy of the proposed method. A flexible marker-net made of foam particles was arranged around the column to provide Lambertian features for the water surface. Two synchronized stereo imaging systems covered all the surrounding areas of the column and provided stereo pair sequences for wave evolution. Subsequently, image segmentation techniques were used to mask the low-confidence disparities in stereo matching, and finally, three-dimensional (3D) wave surfaces were reconstructed in the time sequence. The time histories of the wave elevations at particular locations were extracted and agreed well with the measurements of wave probes with an average bias of 2.4 %. Subsequently, the reconstructed 3D wave field was sliced, exhibiting the instantaneous profiles that agreed with the measurements of wave probes. Moreover, the wave run-up height ratios were consistent with those of a previous study, thereby verifying the method's accuracy from the perspective of spatial evolution. The results demonstrated that the proposed method was capable of precisely measuring the spatial-temporal evolution of the wave field around the square column and displayed potential for application in more studies on wave-structure interactions.

Crack detection in fuel cell electrodes using a spatial filtering technique for overcoming noisy backgrounds

AbstractImage processing is a powerful tool that allows for rapid and automated data parsing in settings that occupy large variable spaces and require large data sets. Feature detection on difficultly discerned backgrounds is a subset of image processing that facilitates the extraction of quantitative metrics from otherwise subjective data. Crack detection and quantification is an important capability in polymer electrolyte membrane fuel cell quality control, failure analysis, and optimization. This work presents a technique to perform crack detection and quantification which overcomes challenges faced by commonly used image segmentation techniques. We demonstrate the use of a geometrically filtered noise‐level detection technique to select a binary threshold value from which we then quantify how cracked a sample is. We demonstrate the accuracy of our technique using programmatically generated test images of known crack amounts and their performance on real‐world fuel cell catalyst layer samples.