Existing Segmentation Algorithms Research Articles

Application of Image Segmentation Algorithms in Computer Vision

In the field of computer vision (CV), image segmentation technology, as a fundamental part, has a crucial impact on the accuracy of subsequent image processing tasks. Image segmentation is not only a crucial transitional step from image processing to image analysis, but also a hot and difficult research topic in the field of CV. Although significant progress has been made in the research of image segmentation algorithms, existing segmentation algorithms may still face challenges in certain specific scenarios due to the complexity and diversity of images, making it difficult to achieve ideal segmentation results. In recent years, the rapid development of deep learning (DL) technology has brought new breakthroughs to the field of image segmentation. DL models, especially Convolutional Neural Networks (CNNs), can capture semantic information of images more accurately by automatically learning feature representations in images, thereby achieving more precise image segmentation. This article delves into the research and application of image segmentation algorithms in CV, with a focus on the application of DL in the field of image segmentation. With the continuous development of advanced technologies such as DL, it is believed that image segmentation technology will play a greater role in more fields in the future.

Human Epithelial Cell Image Analysis and Segmentation using Threshold Based Fusion Technique

The most demanding aspect of digital image processing is segmenting an image efficiently. Cell segmentation or classifying cells in an image is essential while analyzing cell images in medical research, especially in spot diagnosis, cancer cell detection, and live-cell imaging segmentation forms a crucial component. This research examines existing segmentation algorithms and suggests a new segmentation technique that employs image filtering and thresholding. Thresholding is an essential part of image analysis and segmentation. Finally, the segmented image and the FCM (fuzzy C-means) based clustered image are merged. In terms of accuracy, sensitivity, dice-coefficient, and Jaccard-coefficient, the simulation coupled with ground truth data is proven to produce better segmentation outcomes.

Mel2Word: A Text-Based Melody Representation for Symbolic Music Analysis

The purpose of this research is to present a natural language processing-based approach to symbolic music analysis. We propose Mel2Word, a text-based representation including pitch and rhythm information, and a new natural language processing-based melody segmentation algorithm. We first show how to create a melody dictionary using Byte Pair Encoding (BPE), which finds and merges the most frequent pairs that appear in a collection of melodies in a data-driven manner. The dictionary is then used to tokenize or segment a given melody. Utilizing various symbolic melody datasets, we conduct an exploratory analysis and evaluate the classification performance of melody representation models on the MTC-ANN dataset. A comparison with existing segmentation algorithms is also carried out. The result shows that the proposed model significantly improves classification performance in comparison to various melodic features and several existing segmentation algorithms.

Efficient and accurate segmentation of adhesive lamellae in tri-modal microstructure of titanium alloy based on OpenCV

The adhesive characteristics of the tri-modal microstructure in titanium alloys, particularly the adhesion of the lamellar phase that constitutes over 50% of the microstructure, significantly hinder its quantitative characterization, further impeding the establishment of the quantitative relationships between the microstructure and mechanical properties. Addressing this challenge, this study utilized the OpenCV computer vision library to accomplish efficient and precise automated segmentation of the adhesive lamellae in tri-modal microstructure by enhancing the existing segmentation algorithm based on concave point matching. Specifically, by taking the number of the adhesive target after segmentation and Harris corner detection threshold as the optimization goal and variable respectively, and employing the genetic algorithm, the threshold, accounting for the segmentation effect, was adaptively determined to accurately identify the corner points of the adhesive lamellae. And then quick determination of the concave points within these corner points was achieved by substituting conventional pixel circle with equidistant pixels distributed along the circumference. Upon identifying the concave points, classification of the points was performed based on their spacing and location, and corresponding matching criteria were established and applied. Eventually, by introducing segmentation lines between the matched points, adhesive lamellae were segmented with an accuracy exceeding 85%, which will lay a robust foundation for the rapid and automated quantitative characterization of the tri-modal microstructure in titanium alloys.

Filament-necking localization method via combining improved PSO with rotated rectangle algorithm for safflower-picking robots

Safflower is one of the most important specialty economic crops in the world. Safflower blossoms are harvested continuously for 3–5 times. Untimely picking can affect the opening of filaments in the next crop, resulting in reduced filament production and serious economic losses. However, the safflower images collected by picking robots were affected by lighting conditions or complex backgrounds. The existing segmentation algorithms have the problem of insufficient segmentation or over-segmentation, because of the small safflower size and localization area. Therefore, combining improved particle swarm optimization (PSO) with a rotated rectangle algorithm based on the filament-necking localization method for safflower-picking robots was proposed. The R-component image in RGB color space was extracted as a preprocessing sample by combining the color features of safflower. The inertia weights in the PSO algorithm are improved to enhance the performance of the algorithm. An adaptive nonlinear function is introduced to search for the optimal threshold and initially segmented to obtain the binary image. Then, the barycenter and minimum outer rectangle of the contour were set up using the rotated rectangle algorithm based on the geometric features of the filaments. The circular region of interest (Cir-ROI) of filament-necking is determined. The Zhang-Suen refinement algorithm was used for skeleton extraction to design an algorithm for localizing the picking point of safflower filaments. To test safflower images collected in complex environments, the results of average processing time were 0.14 s, and the average relative target area error rate was 19.33 %. Moreover, the localization accuracy of the picking point was 89.75 %. The filament-necking localization method provides a theoretical basis and experimental data support for damage reduction and efficient harvesting of safflower filaments.

Real-Time Segmentation of Artificial Targets Using a Dual-Modal Efficient Attention Fusion Network

The fusion of spectral–polarimetric information can improve the autonomous reconnaissance capability of unmanned aerial vehicles (UAVs) in detecting artificial targets. However, the current spectral and polarization imaging systems typically suffer from low image sampling resolution, which can lead to the loss of target information. Most existing segmentation algorithms neglect the similarities and differences between multimodal features, resulting in reduced accuracy and robustness of the algorithms. To address these challenges, a real-time spectral–polarimetric segmentation algorithm for artificial targets based on an efficient attention fusion network, called ESPFNet (efficient spectral–polarimetric fusion network) is proposed. The network employs a coordination attention bimodal fusion (CABF) module and a complex atrous spatial pyramid pooling (CASPP) module to fuse and enhance low-level and high-level features at different scales from the spectral feature images and the polarization encoded images, effectively achieving the segmentation of artificial targets. Additionally, the introduction of the residual dense block (RDB) module refines feature extraction, further enhancing the network’s ability to classify pixels. In order to test the algorithm’s performance, a spectral–polarimetric image dataset of artificial targets, named SPIAO (spectral–polarimetric image of artificial objects) is constructed, which contains various camouflaged nets and camouflaged plates with different properties. The experimental results on the SPIAO dataset demonstrate that the proposed method accurately detects the artificial targets, achieving a mean intersection-over-union (MIoU) of 80.4%, a mean pixel accuracy (MPA) of 88.1%, and a detection rate of 27.5 frames per second, meeting the real-time requirement. The research has the potential to provide a new multimodal detection technique for enabling autonomous reconnaissance by UAVs in complex scenes.

Automatic detection of microaneurysms using a novel segmentation algorithm based on deep learning techniques

AbstractMicroaneurysms is the first stage of diabetic retinopathy (DR) and it plays a vital role in the computerized diagnosis. However, it is difficult to automatically detect microaneurysms in fundus images due to the complicated background and various illumination reasons. The motivation behind this, is the number of increases in diabetic patients is very large when compared with the number of ophthalmologists. The FSCA‐UNET (Frequency Spatial Channel Attention UNET) segmentation model, is proposed and it is an improvement over UNET. We first use the frequency channel attention mechanism to analyze the features that were extracted from the first stage of the convolution layer, and we obtain good results. Then, we included a spatial attention map with frequency attention, also known as FSCA‐UNET, which makes use of inter‐spatial connections between features. Our deep neural model with an encoder‐decoder structure termed FSCA‐UNET produced more accurate results. Our novel algorithm outdated the performance measures of the existing segmentation algorithms. The proposed segmentation algorithm was trained and tested on Indian Diabetic Retinopathy Image Dataset (IDRiD), and E‐ophtha Dataset and we got promising results in terms of sensitivity, specificity, dice coefficient, precision, F1 score, and accuracy.

A clustering-optimized segmentation algorithm and application on food quality detection

For solving the problem of quality detection in the production and processing of stuffed food, this paper suggests a small neighborhood clustering algorithm to segment the frozen dumpling image on the conveyor belt, which can effectively improve the qualified rate of food quality. This method builds feature vectors by obtaining the image's attribute parameters. The image is segmented by a distance function between categories using a small neighborhood clustering algorithm based on sample feature vectors to calculate the cluster centers. Moreover, this paper gives the selection of optimal segmentation points and sampling rate, calculates the optimal sampling rate, suggests a search method for optimal sampling rate, as well as a validity judgment function for segmentation. Optimized small neighborhood clustering (OSNC) algorithm uses the fast frozen dumpling image as a sample for continuous image target segmentation experiments. The experimental results show the accuracy of defect detection of OSNC algorithm is 95.9%. Compared with other existing segmentation algorithms, OSNC algorithm has stronger anti-interference ability, faster segmentation speed as well as more efficiently saves key information ability. It can effectively improve some disadvantages of other segmentation algorithms.

Mask Positioner: An effective segmentation algorithm for green fruit in complex environment

In order to enable intelligent orchard management and the application of harvesting robots, it is necessary to improve the accuracy of computer vision technology for green fruit segmentation in complex orchard environments. However, existing segmentation algorithms are unable to generate precise fruit masks in such environments. This paper proposes a novel and efficient segmentation algorithm called Mask Positioner for accurate fruit segmentation. The Mask Positioner applies a layer-by-layer filtering approach to refine feature maps generated by the detail refinement network, resulting in a refined mask. The selected pixels are then input to the order decoder to determine their relevance to the fruit region. Finally, the determined pixels are used to generate the final mask, resulting in accurate and efficient fruit segmentation. Mask Positioner is verified by a green persimmon dataset made for the complex background. The experimental results show that the segmentation accuracy of Mask Positioner reaches 67.4%, and the detection accuracy reaches 69.1%. For small fruits, its detection and segmentation accuracy are at least 1.0 and 3.2 percentage points higher than other algorithms. Additionally, the generalization ability of the algorithm is verified using a green apple dataset. Experiments show that it does well in the green fruit segmentation.

Hybrid Multilevel Thresholding Image Segmentation Approach for Brain MRI.

A brain tumor is an abnormal growth of tissues inside the skull that can interfere with the normal functioning of the neurological system and the body, and it is responsible for the deaths of many individuals every year. Magnetic Resonance Imaging (MRI) techniques are widely used for detection of brain cancers. Segmentation of brain MRI is a foundational process with numerous clinical applications in neurology, including quantitative analysis, operational planning, and functional imaging. The segmentation process classifies the pixel values of the image into different groups based on the intensity levels of the pixels and a selected threshold value. The quality of the medical image segmentation extensively depends on the method which selects the threshold values of the image for the segmentation process. The traditional multilevel thresholding methods are computationally expensive since these methods thoroughly search for the best threshold values to maximize the accuracy of the segmentation process. Metaheuristic optimization algorithms are widely used for solving such problems. However, these algorithms suffer from the problem of local optima stagnation and slow convergence speed. In this work, the original Bald Eagle Search (BES) algorithm problems are resolved in the proposed Dynamic Opposite Bald Eagle Search (DOBES) algorithm by employing Dynamic Opposition Learning (DOL) at the initial, as well as exploitation, phases. Using the DOBES algorithm, a hybrid multilevel thresholding image segmentation approach has been developed for MRI image segmentation. The hybrid approach is divided into two phases. In the first phase, the proposed DOBES optimization algorithm is used for the multilevel thresholding. After the selection of the thresholds for the image segmentation, the morphological operations have been utilized in the second phase to remove the unwanted area present in the segmented image. The performance efficiency of the proposed DOBES based multilevel thresholding algorithm with respect to BES has been verified using the five benchmark images. The proposed DOBES based multilevel thresholding algorithm attains higher Peak Signal-to-Noise ratio (PSNR) and Structured Similarity Index Measure (SSIM) value in comparison to the BES algorithm for the benchmark images. Additionally, the proposed hybrid multilevel thresholding segmentation approach has been compared with the existing segmentation algorithms to validate its significance. The results show that the proposed algorithm performs better for tumor segmentation in MRI images as the SSIM value attained using the proposed hybrid segmentation approach is nearer to 1 when compared with ground truth images.

Rayleigh-based segmentation of ISAR images.

Inverse synthetic aperture radar (ISAR) provides a solution to increase the radar angular resolution by observing a moving target over time. The high-resolution ISAR image should undergo a segmentation step to get the target's point cloud data, which is then used for classification purposes. Existing segmentation algorithms seek an optimal threshold in an iterative manner, which adds to the complexity of ISAR and results in an increase in the processing time. In this paper, we take advantage of the distribution of the ISAR image intensity, which is based on the Rayleigh distribution, and obtain an explicit relationship for the optimal segmentation threshold. The proposed segmentation algorithm alleviates the requirement for iterative optimization and its efficiency is shown using both simulated and experimental ISAR images.

Anatomical Feature Segmentation of Femur Point Cloud Based on Medical Semantics

Feature segmentation is an essential phase for geometric modeling and shape processing in anatomical study of human skeleton and clinical digital treatment of orthopedics. Due to various degrees of freedom of bone surface, the existing segmentation algorithms can hardly meet specific medical need. To address this, a novel segmentation methodology for anatomical features of femur model based on medical semantics is put forward. First, anatomical reference objects (ARO) are created to represent typical characteristics of femur anatomy by 3D point fitting in combination with medical priori knowledge. Then, local point clouds between adjacent anatomies are selected according to the AROs to extract boundary feature point (BFP)s. Finally, the complete model of femur is divided into anatomical regions by executing the enhanced watershed algorithm guided with BFPs. Experimental results show that the proposed method has the advantages of automatic segmentation of femoral head, neck and other complex areas, and the segmentation results have better medical semantics. In addition, the slight modification of segmentation results can be achieved by adjusting a few threshold parameter values, which improves the convenience of modification for ordinary users.

Automatic segmentation of ultrasound images using SegNet and local Nakagami distribution fitting model

Automatic and accurate segmentation of ultrasound (US) images plays a vital role in computer-aided diagnosis of many diseases. However, multiple degradations within US images, such as speckle noise, intensity inhomogeneity, shadows, low contrast, and low signal-to-noise ratio, often cover up the details of imaging tissues and blur the edges of lesions, resulting in limited accuracy for existing segmentation algorithms. To tackle these issues, a coarse-to-fine segmentation method combining deep learning with an active contour model (ACM) is proposed in this paper. At the first stage, a SegNet is trained and employed to predict the segmentation mask because it not only has a superior boundary delineation capability, but also has high efficiency in terms of memory and computational time during inference. Due to the multiple degradations of US images, the unavailability of large-scale US image datasets, and the loss of spatial information during downsampling, the SegNet may only infer the coarse object boundaries, especially for the tissue with blurred edges. At the second stage, the segmentation mask of the SegNet is refined to locate the accurate object boundaries by a novel ACM termed the local Nakagami distribution fitting (LNDF) model, which takes advantage of discrepancies between Nakagami distributions of different tissues around the boundaries within US images. Experimental results on two US image datasets demonstrate that, compared with some state-of-the-art segmentation methods, our proposed method achieves the highest segmentation accuracy at the cost of acceptable running time, and thereby adapts to the automatic segmentation of US images in clinical applications.

CNAViz: An interactive webtool for user-guided segmentation of tumor DNA sequencing data.

Copy-number aberrations (CNAs) are genetic alterations that amplify or delete the number of copies of large genomic segments. Although they are ubiquitous in cancer and, thus, a critical area of current cancer research, CNA identification from DNA sequencing data is challenging because it requires partitioning of the genome into complex segments with the same copy-number states that may not be contiguous. Existing segmentation algorithms address these challenges either by leveraging the local information among neighboring genomic regions, or by globally grouping genomic regions that are affected by similar CNAs across the entire genome. However, both approaches have limitations: overclustering in the case of local segmentation, or the omission of clusters corresponding to focal CNAs in the case of global segmentation. Importantly, inaccurate segmentation will lead to inaccurate identification of CNAs. For this reason, most pan-cancer research studies rely on manual procedures of quality control and anomaly correction. To improve copy-number segmentation, we introduce CNAViz, a web-based tool that enables the user to simultaneously perform local and global segmentation, thus overcoming the limitations of each approach. Using simulated data, we demonstrate that by several metrics, CNAViz allows the user to obtain more accurate segmentation relative to existing local and global segmentation methods. Moreover, we analyze six bulk DNA sequencing samples from three breast cancer patients. By validating with parallel single-cell DNA sequencing data from the same samples, we show that by using CNAViz, our user was able to obtain more accurate segmentation and improved accuracy in downstream copy-number calling.

Reducing the Presence of Clusters in Bubble Size Measurements for Gas Dispersion Characterizations

This short communication evaluates a new strategy to sample bubbles in gas dispersion characterizations. Bubble size is measured in a bidimensional flotation cell using the McGill bubble size analyzer under different types of frothers, frother concentrations and superficial gas rates. The original design of this bubble viewer is modified, changing the deflecting system to photograph only a fraction of the bubbles entering the device. As a result, the new design increases the ability to successfully identify bubbles by a maximum of 20% using an automated algorithm. This increase is caused by a reduction in the formation of clusters in the visual field. The improvement, which is a function of the operating conditions, is most significant in the transition from ellipsoidal/ellipsoidal–turbulent regimes (no frother or low frother concentrations) to conditions with an over-agglomeration of bubbles in the visual field (high superficial gas rates and high frother concentrations). A comparison of the bubble size parameters obtained from the original and proposed deflecting systems shows that the new design does not distort the estimated bubble size distributions. To complement the research findings, alternative sampling designs, using new or existing segmentation algorithms, are then proposed to improve gas dispersion characterizations at different scales.

Multi-Scale Semantic Segmentation for Fire Smoke Image Based on Global Information and U-Net

Smoke is translucent and irregular, resulting in a very complex mix between background and smoke. Thin or small smoke is visually inconspicuous, and its boundary is often blurred. Therefore, it is a very difficult task to completely segment smoke from images. To solve the above issues, a multi-scale semantic segmentation for fire smoke based on global information and U-Net is proposed. This algorithm uses multi-scale residual group attention (MRGA) combined with U-Net to extract multi-scale smoke features, and enhance the perception of small-scale smoke. The encoder Transformer was used to extract global information, and improve accuracy for thin smoke at the edge of images. Finally, the proposed algorithm was tested on smoke dataset, and achieves 91.83% mIoU. Compared with existing segmentation algorithms, mIoU is improved by 2.87%, and mPA is improved by 3.42%. Thus, it is a segmentation algorithm for fire smoke with higher accuracy.

Classification of myocardial fibrosis in DE-MRI based on semi-supervised semantic segmentation and dual attention mechanism

ObjectiveIt is essential to utilize cardiac delayed-enhanced magnetic resonance imaging (DE-MRI) to diagnose cardiovascular disease. By segmenting myocardium DE-MRI images, it provides critical information for the evaluation and treatment of myocardial infarction. As a consequence, it is vital to investigate the segmentation and classification technique of myocardial DE-MRI. MethodsFirstly, an end-to-end minimally supervised and semi-supervised semantic DE-MRI myocardial fibrosis segmentation framework is proposed, which combines image classification and semantic segmentation branches based on the self-attention mechanism. Following that, a residual hole network fused with the dual attention mechanism was built, and a double attention metabolic pathway classification method for cardiac fibrosis in DE-MRI images was developed. ResultsBy adding pixel-level labels to an extra 40 training images, the segmentation model may enhance semantic segmentation performance by 2.6 percent (from 61.2 percent to 63.8 percent). When the number of pixel-level labels is increased to 80, semi-supervised feature extraction increases by 4.7 percent when compared to weakly guided semantic segmentation. Adding an attention mechanism to the critical network DRN (Deep Residual Network) can increase the classifier's performance by a small amount. Experiments revealed that the models worked effectively. ConclusionThis paper investigates the segmentation and classification of cardiac fibrosis in DE-MRI data using a semi-supervised semantic segmentation and dual attention mechanism, dealing with the issue that existing segmentation algorithms have difficulty segmenting myocardial fibrosis tissue. In the future, we can consider optimizing the design of the attention module to reduce the module computation.

COMPARITIVE ANALYSIS OF K++GKFCM CLUSTERING FOR BRAIN TUMOUR DETECTION

Due to the complicated structure of brain tumors, hazy borders, and outside variables like noise, inferring tumor and edema regions from brain magnetic resonance imaging (MRI) data is still difficult. In this paper, a powerful hybrid clustering technique together with morphological procedures is suggested for segmenting brain tumors in order to reduce noise sensitivity and enhance segmentation stability. The following are the paper's key contributions: initially, adaptive Wiener filtering is employed for de-noising, and morphological procedures are applied for deleting non-brain tissue, thereby minimizing the method’s susceptibility to noise. Second, to segment pictures, the fuzzy C-means technique based on a Gaussian kernel is used with K-means++ clustering. This clustering decreases the sensitivity of the clustering parameters while simultaneously enhancing the stability of the algorithm. To produce accurate representations of brain tumours, the retrieved tumour pictures are lastly post processed utilising morphological procedures and median filtering. The suggested approach was also contrasted with other existing segmentation algorithms.

Acute lymphoblastic leukemia prediction using covering-oriented rough k-means segmentation algorithm

Acute Lymphoblastic Leukemia (ALL) is a type of dreadful cancer that has an effect on the bone marrow and blood. In recent decades, leukemia mainly affects youngsters and grown-ups. Timely diagnosis of leukemia is crucial to give right treatment to patients, predominantly in the case of children. Computational tecniques for medical image analysis have incredible use and become the focus of research in medical image processing. The major objective of this research work is to predict the acute lymphoblastic leukemia cells using the proposed segmentation algorithms with the assistance of machine learning classifiers. In this study, hybrid clustering based segmentation methods Covering-orientedRough K-Means (CRKM) algorithm is proposed to segment the image of the leukemia nucleus. Then, the performance of the segmentation algorithms were analyzed using different machine learning classifiers. From the experimental analysis, it is observed that the machine learning classification algorithms achieved better prediction accuracy when compared to existing segmentation algorithms.

Apple Sweetness Measurement and Fruit Disease Prediction Using Image Processing Techniques Based on Human-Computer Interaction for Industry 4.0

When it comes to agricultural sciences, one of the most difficult challenges to solve is the detection of diseases. Agricultural specialists study a variety of sources to detect plant issues on a regular basis. Rarely can misinterpretations of diseased plants cause improper pesticide selection and subsequent agricultural disaster, although this does happen from time to time. In order to diagnose illnesses at an early stage, it is necessary to deploy automated disease detection systems. This is critical for farmers since it is both time-consuming and expensive. A sick leaf must be carefully segmented in order to be properly separate it from the rest of the leaves. Despite digital noise, a different background, a different shape, and a different brightness, it is tough to distinguish a sick photo. In order to increase the quality of apple leaf images for disease detection and classification, a new approach known as brightness preserving dynamic fuzzy histogram equalisation (BPDFHE) has been created. To determine the sweetness of an apple, examine the leaf and the texture of the fruit. In the next section, the performance of the proposed enhancement algorithm is compared to the performance of existing enhancement approaches. Existing segmentation algorithms are outperformed by our approach for segmenting the area of interest from ill leaves against a live background. It is during this phase that we analyse the Jaccard index, the Dice coefficient, and correctness. Comparing the proposed segmentation algorithm to current approaches, it proves to be a highly effective strategy that can more efficiently identify apple ill leaves from a live background with a 99.8 percent accuracy rate.