Search

Capsule endoscopy (CE) is a widely used medical imaging tool for the diagnosis of gastrointestinal tract abnormalities like bleeding. However, CE captures a huge number of image frames, constituting a time-consuming and tedious task for medical experts to manually inspect. To address this issue, researchers have focused on computer-aided bleeding detection systems to automatically identify bleeding in real time. This paper presents a systematic review of the available state-of-the-art computer-aided bleeding detection algorithms for capsule endoscopy. The review was carried out by searching five different repositories (Scopus, PubMed, IEEE Xplore, ACM Digital Library, and ScienceDirect) for all original publications on computer-aided bleeding detection published between 2001 and 2023. The Preferred Reporting Items for Systematic Review and Meta-Analyses (PRISMA) methodology was used to perform the review, and 147 full texts of scientific papers were reviewed. The contributions of this paper are: (I) a taxonomy for computer-aided bleeding detection algorithms for capsule endoscopy is identified; (II) the available state-of-the-art computer-aided bleeding detection algorithms, including various color spaces (RGB, HSV, etc.), feature extraction techniques, and classifiers, are discussed; and (III) the most effective algorithms for practical use are identified. Finally, the paper is concluded by providing future direction for computer-aided bleeding detection research.

As a fundamental but difficult topic in computer vision, 3D object segmentation has various applications in medical image analysis, autonomous vehicles, robotics, virtual reality, lithium battery image analysis, etc. In the past, 3D segmentation was performed using hand-made features and design techniques, but these techniques could not generalize to vast amounts of data or reach acceptable accuracy. Deep learning techniques have lately emerged as the preferred method for 3D segmentation jobs as a result of their extraordinary performance in 2D computer vision. Our proposed method used a CNN-based architecture called 3D UNET, which is inspired by the famous 2D UNET that has been used to segment volumetric image data. To see the internal changes of composite materials, for instance, in a lithium battery image, it is necessary to see the flow of different materials and follow the directions analyzing the inside properties. In this paper, a combination of 3D UNET and VGG19 has been used to conduct a multiclass segmentation of publicly available sandstone datasets to analyze their microstructures using image data based on four different objects in the samples of volumetric data. In our image sample, there are a total of 448 2D images, which are then aggregated as one 3D volume to examine the 3D volumetric data. The solution involves the segmentation of each object in the volume data and further analysis of each object to find its average size, area percentage, total area, etc. The open-source image processing package IMAGEJ is used for further analysis of individual particles. In this study, it was demonstrated that convolutional neural networks can be trained to recognize sandstone microstructure traits with an accuracy of 96.78% and an IOU of 91.12%. According to our knowledge, many prior works have applied 3D UNET for segmentation, but very few papers extend it further to show the details of particles in the sample. The proposed solution offers a computational insight for real-time implementation and is discovered to be superior to the current state-of-the-art methods. The result has importance for the creation of an approximately similar model for the microstructural analysis of volumetric data.

Lane marking recognition is one of the most crucial features for automotive vehicles as it is one of the most fundamental requirements of all the autonomy features of Advanced Driver Assistance Systems (ADAS). Researchers have recently made promising improvements in the application of Lane Marking Detection (LMD). This research article has taken the initiative to review lane marking detection, mainly using deep learning techniques. This paper initially discusses the introduction of lane marking detection approaches using deep neural networks and conventional techniques. Lane marking detection frameworks can be categorized into single-stage and two-stage architectures. This paper elaborates on the network’s architecture and the loss function for improving the performance based on the categories. The network’s architecture is divided into object detection, classification, and segmentation, and each is discussed, including their contributions and limitations. There is also a brief indication of the simplification and optimization of the network for simplifying the architecture. Additionally, comparative performance results with a visualization of the final output of five existing techniques is elaborated. Finally, this review is concluded by pointing to particular challenges in lane marking detection, such as generalization problems and computational complexity. There is also a brief future direction for solving the issues, for instance, efficient neural network, Meta, and unsupervised learning.

Background: Lane detection is a difficult issue because of different lane circumstances. It plays an important part in advanced driver assistance systems, which give information about the centre of a host vehicle such as lane structure and lane position. Lane departure warning (LDW) is used to warn the driver about an unplanned lane exit from the original lane. The objective of this study was to develop a data-fusion LDW framework to improve the rate of detection of lane departure during daylight and at night. Methods: Vision-based LDW is a comprehensive framework based on vision-based lane detection with additional lateral offset ratio computations based on the detected X 12 and X 22 coordinates. The computed lateral offset ratio is used to detect lane departure based on predefined LDW identification criteria for vision-based LDW. Data fusion-based LDW was developed using a multi-input-single-output fuzzy logic controller. Data fusion involved lateral offset ratio and yaw acceleration response from the vision-based LDW and model-based vehicle dynamics frameworks. Real-life datasets were generated for simulation under the MATLAB Simulink platform. Results: Experimental results showed that fusion-based LDW achieved an average lane departure detection rate of 99.96% and 98.95% with false positive rates (FPR) of 0.04% and 1.05% using road footage clips #5–#27 in daytime and night-time, respectively. The average FPR using data fusion-based LDW reduced by 18.83% and 15.22% compared to vision-based LDW in daytime and night-time, respectively. Conclusions: The data fusion-based LDW is a novel way of reducing false lane departure detection by fusing two types of modalities to determine the correct lane departure information. The limitation is the constant warning threshold value used in the current implementation of LDW in the vision-based LDW framework. An adaptive mechanism of warning threshold taking various road structures into account could be developed to improve lane departure detection.

Lane departure collisions have contributed into the traffic accidents that cause millions of injuries and tens of thousands of casualties per year worldwide. Hence, a vision-based lane detection framework (VBLD) is proposed to detect lane markings on the road for unindented lane departure. The proposed VBLD framework is composed of colour space conversion, region of interest, lane marking segmentation, Hough transformation and peak detection, reverse Hough transformation, and draw detected lines on original image. Besides, finite impulse response saturation autothreshold (FIRSA) lane marking segmentation method is also proposed for lane edges extraction. For performance evaluation on the proposed VBLD framework and proposed FIRSA lane markings segmentation method, real-life datasets of road footages are collected using an instrumented vehicle. The outputs of lane detection frames from Clip #1, #2, #3, and #4 involving variety of road conditions are evaluated using detection rate, false positive rate, and false negative rate assessment metrics where the number of frames are manually counted using visual inspection. Experimental results have shown the evidences of the proposed VBLD framework using proposed FIRSA lane markings segmentation method obtained satisfactory lane detection results compared to benchmark lane marking segmentation methods. [ABSTRACT FROM AUTHOR]