Region Of Interest Extraction Method Research Articles

A ROI Extraction Method for Wrist Imaging Applied in Smart Bone-Age Assessment System.

Bone Age (BA) is reckoned to be closely associated with the growth and development of teenagers, whose assessment highly depends on the accurate extraction of the reference bone from the carpal bone. Being uncertain in its proportion and irregular in its shape, wrong judgment and poor average extraction accuracy of the reference bone will no doubt lower the accuracy of Bone Age Assessment (BAA). In recent years, machine learning and data mining are widely embraced in smart healthcare systems. Using these two instruments, this article aims to tackle the aforementioned problems by proposing a Region of Interest (ROI) extraction method for wrist X-ray images based on optimized YOLO model. The method combines Deformable convolution-focus (Dc-focus), Coordinate attention (Ca) module, Feature level expansion, and Efficient Intersection over Union (EIoU) loss all together as YOLO-DCFE. With the improvement, the model can better extract the features of irregular reference bone and reduce the potential misdiscrimination between the reference bone and other similarly shaped reference bones, improving the detection accuracy. We select 10041 images taken by professional medical cameras as the dataset to test the performance of YOLO-DCFE. Statistics show the advantages of YOLO-DCFE in detection speed and high accuracy. The detection accuracy of all ROIs is 99.8%, which is higher than other models. Meanwhile, YOLO-DCFE is the fastest of all comparison models, with the Frames Per Second (FPS) reaching 16.

Region of interest (ROI) extraction and crack detection for UAV-based bridge inspection using point cloud segmentation and 3D-to-2D projection

For digital-image-based bridge inspection tasks, images captured by camera-carrying unmanned aircraft vehicles (UAVs) usually contain both the region of interest (ROI) and the background. However, accurately detecting cracks in concrete surface images containing background information is challenging. To improve UAV-based bridge inspection, an image extraction and crack detection methodology is presented in this paper. First, a deep-learning-based semantic segmentation network RandLA-BridgeNet for large-scale bridge point clouds, which can facilitate 3D ROI extraction, is trained and tested. Second, an image ROI extraction method based on 3D-to-2D projection is presented to generate images containing only the ROI. Finally, a data-driven deep learning convolutional neural network (CNN) called the grid-based classification and box-based detection fusion model (GCBD) is utilized to identify cracks in the processed images. An experiment is conducted on highway bridge images to validate the presented methodology. The overall semantic segmentation and image ROI extraction accuracies are 97.0% and 98.9%, respectively. After ROI extraction, 47.9% of the grid cells, which represent background misrecognition, are filtered, greatly improving the crack identification accuracy.

An Inspection Technique Using Fit Clearance Based on Microscopic Vision in Precision Assembly.

Inspection is a crucial process to ensure product quality. In the precision assembly of an optic-mechanical device, a part with micro multi-section arcs needs to be inspected and assembled into another part. Actually, because of machining errors, including dimensional and geometric shapes, can lead to complex deformation modes for parts with micro multi-section arcs, posing challenges to their inspection. Furthermore, inconsistencies in feature images in microscopic vision may complicate the extraction of the Region of Interest (ROI). To address these issues, this paper proposes an ROI extraction method based on the CAD model for rough positioning of feature points and connected region detection for refinement. Subsequently, based on feature points, the CAD model is used again to obtain the ROI. For inspection purposes, this paper proposes a method suitable for micro multi-section arcs based on assembly fit requirements. Experimental testing was performed on parts with eight-section arcs and mirrors to verify the effectiveness of the proposed method. This method provides a suitable solution for the inspection of micro multi-section arcs in precision assembly with the potential to improve the accuracy of the inspection results.

AR-DIC: A Novel Automatic Optical Strategy for Displacement Measurement of the Curved-in-Place Pipe Under Dynamic Fracturing

Curved-In-Place Pipe (CIPP) technology is widely used in urban underground pipeline repair nowadays. However, the lack of automatic global displacement acquisition method makes it still quite challenging to effectively obtain the detailed displacement characteristics of the CIPP pipes under dynamic loads. Herein, we propose a cascade strategy named automatic speckle region extraction digital image correlation (AR-DIC) to achieve the dynamic, efficient and accurate displacement estimation of the CIPP pipelines, so as to determine the variation characters of the CIPP pipelines under dynamic loads. In the first stage, an improved automatic DeepLabv3+ neural network (IA-DeepLabv3+) is meticulously designed to extract the region of interest (ROI) of the CIPP pipeline speckle automatically. A dataset containing the cluttered background and the real experimental speckle are also generated to train the model parameters. Note that this is the first time an automated ROI extraction method has been proposed in CIPP, even in the DIC field. In the second stage, the digital image correlation based on scale-invariant feature transform (SIFT-DIC) algorithm is adopted to calculate the extracted ROI and get the displacement field. Finally, the proposed method is tested on both the simulated and experimental datasets. The experimental results confirm that our method can not only extract the speckle ROI region accurately, but also achieve high-precision calculation of the displacement field.

A Contactless Palmprint Imaging System Design Using Mediapipe Hands

Palmprint has been widely used in biometric systems because of its durability and reliability. To avoid recognition performance degradation, dynamic region of interest extraction is a critical step for these systems. In this study, a low-cost contactless palmprint imaging system has been designed and a dynamic region of interest extraction method has been applied to palmprints using the MediaPipe Hands framework. Since the need for hygienic touchless systems has been realized in the post-COVID-19 pandemic world, a low-cost imaging system has been proposed to capture the user’s hand at a distance without touching any platform. The region of interest of the user's palmprints in a real-time video stream has been extracted dynamically. This study creates a paradigm for future studies on palmprint imaging. With conducted experiments, the potential of MediaPipe Hands in terms of speed and accuracy on mobile palmprint imaging applications has been realized on Raspberry Pi 4. This work demonstrates that the employed hardware and proposed hand-tracking algorithm are suitable for designing low-cost contactless palmprint imaging systems in non-controlled ambient light conditions. For recognition purposes, a database will be released soon.

Palmprint Phenotype Feature Extraction and Classification Based on Deep Learning.

Palmprints are of long practical and cultural interest. Palmprint principal lines, also called primary palmar lines, are one of the most dominant palmprint features and do not change over the lifespan. The existing methods utilize filters and edge detection operators to get the principal lines from the palm region of interest (ROI), but can not distinguish the principal lines from fine wrinkles. This paper proposes a novel deep-learning architecture to extract palmprint principal lines, which could greatly reduce the influence of fine wrinkles, and classify palmprint phenotypes further from 2D palmprint images. This architecture includes three modules, ROI extraction module (REM) using pre-trained hand key point location model, principal line extraction module (PLEM) using deep edge detection model, and phenotype classifier (PC) based on ResNet34 network. Compared with the current ROI extraction method, our extraction is competitive with a success rate of 95.2%. For principal line extraction, the similarity score between our extracted lines and ground truth palmprint lines achieves 0.813. And the proposed architecture achieves a phenotype classification accuracy of 95.7% based on our self-built palmprint dataset CAS_Palm.

An ROI Extraction Method of Finger Vein Images Based on Large Receptive Field Gradient Operator for Accurate Localization of Joint Cavity.

Region of interest (ROI) extraction is a key step in finger vein recognition preprocessing. The current method takes the finger region in the vein image as the ROI, but this method cannot obtain better recognition accuracy because it only removes the background noise of the image and ignores factors such as the position and shape of the finger. To solve this problem, we limited the ROI to a fixed region between two finger joint cavities, proposed a new large receptive field gradient operator, and designed and implemented a new method for ROI extraction. It uses a large receptive field to search the target, which is similar to human vision, thus solving the problem of difficult ROI localization for images with large gradient areas. Moreover, for external factors such as noise and uneven illumination in the vein image, the interference factors can be eliminated by averaging them to a larger range with a larger size operator to improve the accuracy of the subsequent matching recognition. To verify the effectiveness of the proposed method, we conducted sufficient matching experiments on three public finger vein datasets. On various datasets, our method significantly reduced the identified EER value, with the lowest EER value reaching 0.96%. The experimental results show that the proposed ROI extraction method can effectively eliminate the influence of finger position, finger shape, and other factors on the subsequent recognition performance, accurately locate the finger joint cavity, and effectively improve the recognition performance.

V-Vibe: A robust ROI extraction method based on background subtraction for vein images collected by infrared device

In recent years, the global epidemic has put forward higher requirements for public health security, and the non-contact finger vein acquisition and identification technology based on infrared imaging has become a research hotspot due to its high security, accuracy and convenience. However, infrared images generally have the problems of low resolution, low contrast and low signal-to-noise ratio, which seriously affect the effective extraction and utilization of image feature information. Moreover, most of the images captured by unconstrained vein acquisition devices have a large number of complex backgrounds, which are often similar to the pixel distribution of finger region, which greatly affects the recognition performance of vein images. To solve the above problems, we carried out the following exploratory work. First, we designed and developed a new finger vein infrared acquisition device. The device adopts a kind of physical structure of triangular spotlight, and in combination with our proposed V-Vibe method can achieve complete Region of Interest (ROI) extraction of finger vein images. Second, we created a new finger vein dataset (named FV-BS) for background subtraction. The dataset contains 420 individuals, 4,200 finger vein image samples and 420 device background images, as well as a subset of vein image data filled with complex backgrounds. Third, we proposed a new method named V-Vibe to extract ROI from vein images by using background subtraction technology. Subtraction can effectively eliminate factors that affect recognition performance in both vein images and background sample set, thus greatly reducing the influence of complex backgrounds and low-quality infrared images on identification. To verify the validity and accuracy of the V-Vibe method, we designed and conducted a large number of experiments on several datasets, including four publicly available datasets, as well as self-built FV-BS, and analyzed the results in detail. Compared with other methods, the ROI extraction effect of V-Vibe method proposed in this paper is significantly improved, and EER is reduced by 1.18% and 2.37% on average respectively in maximum curvature and repeated line tracking feature extraction methods, which is an effective method for ROI extraction of finger vein images. Furthermore, we also tested the proposed V-Vibe method to extract ROI from the vein images of palm and back of hands. Experimental results show that our method has excellent segmentation accuracy on both palm and dorsal hand vein datasets, with the average F1 Score reaching 95.1%.

A new scheme of vehicle detection for severe weather based on multi-sensor fusion

Automated vehicles are prone to traffic accidents in severe weather conditions. Real-time vehicle detection can improve the driving safety of automated vehicles. This paper proposes a new vehicle detection method based on multi-sensor fusion to improve the vehicle detection performance in severe weather conditions. First, an efficient vehicle target extraction method from the radar is proposed that uses supervised learning to train a classifier based on LightGBM. This method does not require complex prior knowledge to determine the target segmentation threshold and transforms the target extraction into a data-driven classification. The vehicle target extraction method based on LightGBM has 95.5% accuracy and a 96% true positive rate. Second, we estimate the potential area of vehicles from infrared images according to the distribution of radar targets and predict the region of interest (ROI) of vehicles based on pixel regression. The ROI extraction method based on radar can avoid complicated calculations and interference of heat sources in the environment, which will greatly improve the speed and accuracy of ROI extraction. Radar-based ROI extraction only takes 4 ms, which is much lower than image-based ROI extraction. Finally, four new Haar-like feature templates are designed to improve the vehicle detection performance, which can improve the detection accuracy by 2.9%. This method has a 92.4% detection accuracy and a 43 Fps detection speed in the road test, which significantly improves the vehicle detection performance in severe weather.

Application of region of interest extraction method based on deep learning in UAV high performance image compression

UAV has been widely used in detecting targets, but the image transmission of UAV still faces the problems of distortion and frame loss. Most image compression methods based on deep learning are lossy compression, and lossy compression reduces image quality in exchange for higher compression ratio. In order to improve the quality of the region of interest (ROI) in the reconstructed image with a certain bit rate, an importance map extraction module is embedded in the encoder, and the importance map is generated by extracting the output features of the last layer of the encoder. Finally, the mask is generated to guide the efficient allocation of bit rate in the process of drop coding. At the same time, a decoder enhancement module is embedded in the decoder output to predict the high frequency components in the reconstructed image and improve the quality of the reconstructed image by enhancing the details in the reconstructed image. The experimental results show that the proposed method is superior to the comparison method when multi-scale structural similarity (MS-SSIM) is used as the evaluation index, and the proposed method achieves better visual perception quality.

A New Adaptive Region of Interest Extraction Method for Two-Lane Detection

A New Adaptive Region of Interest Extraction Method for Two-Lane Detection

A Flexible Region of Interest Extraction Algorithm with Adaptive Threshold for 3-D Synthetic Aperture Radar Images

Most of the existing image segmentation methods have a strong anti-noise ability but are susceptible to the interference in the background, so they are not suitable for 3-D synthetic aperture radar (SAR) image target extraction. Region of interest (ROI) extraction can improve the anti-interference ability of the image segmentation methods. However, the existing ROI extraction method uses the same threshold to process all the images in the data set. This threshold is not optimal for each image. Designed for 3-D SAR image target extraction, we propose an ROI extraction algorithm with adaptive threshold (REAT) to enhance the anti-interference ability of the existing image segmentation methods. The required thresholds in the proposed algorithm are adaptively obtained by the mapping of the image features. Moreover, the proposed algorithm can easily be applied to existing image segmentation methods. The experiments demonstrate that the proposed algorithm significantly enhances the anti-interference ability and computational efficiency of the image segmentation methods. Compared with the existing ROI extraction algorithm, the proposed algorithm improves the dice similarity coefficient by 6.4%.

A Novel ROI Extraction Method Based on the Characteristics of the Original Finger Vein Image.

As the second generation of biometric technology, finger vein recognition has become a research hotspot due to its advantages such as high security, and living body recognition. In recent years, the global pandemic has promoted the development of contactless identification. However, the unconstrained finger vein acquisition process will introduce more uneven illumination, finger image deformation, and some other factors that may affect the recognition, so it puts forward higher requirements for the acquisition speed, accuracy and other performance. Considering the universal, obvious, and stable characteristics of the original finger vein imaging, we proposed a new Region Of Interest (ROI) extraction method based on the characteristics of finger vein image, which contains three innovative elements: a horizontal Sobel operator with additional weights; an edge detection method based on finger contour imaging characteristics; a gradient detection operator based on large receptive field. The proposed methods were evaluated and compared with some representative methods by using four different public datasets of finger veins. The experimental results show that, compared with the existing representative methods, our proposed ROI extraction method is th of the processing time of the threshold-based methods, and it is similar to the time spent for coarse extraction in the mask-based methods. The ROI extraction results show that the proposed method has better robustness for different quality images. Moreover, the results of recognition matching experiments on different datasets indicate that our method achieves the best Equal Error Rate (EER) of without the refinement of feature extraction parameters, and all the EERs are significantly lower than those of the representative methods.

A Camera and LiDAR Data Fusion Method for Railway Object Detection

Object detection on railway tracks, which is crucial for train operational safety, face numerous challenges such as multiple types of objects and the complexity of train running environment. In this study, a multi-sensor framework is proposed to fuse camera and LiDAR data for the detection of objects on railway track including small obstacles and forward trains. The framework involves a two-stage process: region of interest extraction and object detection. In the first stage, a multi-scale prediction network is designed to achieve pixel level segmentation of the railway track and forward train via the image. In the second stage, LiDAR data is used to estimate the distance to the train and detect small obstacles in the railway track area which is extracted from the first stage. Experimental results show that the region of interest extraction method achieves desirable accuracy for railway track and train segmentation; and the proposed fusion method outperforms the one based on camera or LiDAR alone for small obstacles and forward train detection. Moreover, in practice the proposed framework has been successfully applied on the Hong Kong Metro TSUEN WAN line and the Beijing Metro YANFANG line.

Traffic vehicle cognition in severe weather based on radar and infrared thermal camera fusion

The accuracy and robustness of vehicle cognition in severe weather have always been the focus and difficulty of intelligent vehicle environment perception. This paper proposes a vehicle cognition method based on radar and infrared thermal camera information fusion in severe weather. The fusion of radar and infrared cameras can greatly enrich the completeness of vehicle cognitive information. First, an attention mechanism based on radar guidance information was proposed to extract the vehicle region of interest (ROI). Unlike the traditional ROI extraction method, this method is not disturbed by the environment and does not need complicated calculations, which can quickly and accurately extract ROI for vehicles. Second, based on ROI information, the infrared thermal image is reconstructed and enhanced, which is of great significance to improve the accuracy of vehicle detection for deep learning. Finally, we propose a vehicle depth estimation method based on pixel regression and use multi-scale cognitive information to fuse radar and image targets. The fusion method considering depth information can reduce target confusion and improve fusion robustness and accuracy, especially when vehicles are adjacent to each other. The experimental results show that the vehicle detection accuracy of this method is 95.2% and the vehicle detection speed is 37 fps, which effectively improves the performance of vehicle detection in severe weather.

Enhanced Region of Interest Extraction method for Finger Vein Images

Extracting Region of Interest (ROI) is an important step for finger vein recognition system. The purpose of this process is to determine the part of the image that we need for extracting features. In this paper we present an ROI extraction method that overcome the problems of finger rotation and displacement. We first locate the finger midline to be used in correcting the oblique images. We then use a sliding window to determine the Proximal inter phalangeal joint and to further identify the ROI height. Finally, from the corrected image of a certain height, the ROI is obtained through the use of finger edges internal tangents as ROI boundaries. The results prove that our method in a more accurate and effective manner in comparison with the method of [1], and thus enhance the performance of the system.

A review of mammographic region of interest classification

AbstractEarly detection of breast cancer is important and highly valuable in clinical practice. X‐ray mammography is broadly used for prescreening the breast and is also attractive due to its noninvasive nature. However, experts can misdiagnose a significant proportion of the cases, which may either cause redundant examinations or cancer. In order to reduce false positive and negative rates of mammography screening, computer‐aided breast cancer detection has been studied for more than 30 years and many methods have been proposed by the researchers. In this review, region of interest (ROI) classification methods, which operate on a predefined or segmented ROIs with a focus on mass classification are surveyed. A total of 72 high quality journal and conference papers are selected from the Web of Science (WOS) database that meet several inclusion criteria. A comparative analysis is provided based on ROI extraction methods, data sets and machine learning techniques employed, the prediction accuracies, and usage frequency statistics. Based on the performances obtained on publicly available data sets, the ROI classification problem from mammogram images can be considered as approaching to be solved. Nonetheless, it can still be used as complementary information in breast cancer detection from the whole mammograms, which has room for improvement.This article is categorized under: Application Areas > Science and Technology Technologies > Machine Learning Technologies > Classification

Toward Unconstrained Palmprint Recognition on Consumer Devices: A Literature Review

As a biometric palmprints have been largely under-utilized, but they offer some advantages over fingerprints and facial biometrics. Recent improvements in imaging capabilities on handheld and wearable consumer devices have re-awakened interest in the use fo palmprints. The aim of this paper is to provide a comprehensive review of state-of-the-art methods for palmprint recognition including Region of Interest extraction methods, feature extraction approaches and matching algorithms along with overview of available palmprint datasets in order to understand the latest trends and research dynamics in the palmprint recognition field.

Efficient Region of Interest Extraction Methods for Multiple Medical Images

Medical images do contain important and unimportant spatial regions. Compression methods which are capable of reconstructing the image with high quality are required to compress the medical images. For these images, only a portion of it is useful for diagnosis hence a region based coding techniques are significant for compressing and transmission. Extracting a significant region is of great demand since a slighter mistake may leads to wrong diagnosis. This paper is focused on investigating multiple image processing algorithms for medical images. All the images may not contain the same region of interest, so different approaches are supposed to apply for different images. In this three types of medical images were considered like magnetic resonance (MR) brain images, computer tomography (CT) abdomen images and X-ray lung images. In this paper three automatic region of interest extraction algorithms were proposed for different types of images.

Convolutional neural network based detection and judgement of environmental obstacle in vehicle operation

Precise real-time obstacle recognition is both vital to vehicle automation and extremely resource intensive. Current deep-learning based recognition techniques generally reach high recognition accuracy, but require extensive processing power. This study proposes a region of interest extraction method based on the maximum difference method and morphology, and a target recognition solution created with a deep convolutional neural network. In the proposed solution, the central processing unit and graphics processing unit work collaboratively. Compared with traditional deep learning solutions, the proposed solution decreases the complexity of algorithm, and improves both calculation efficiency and recognition accuracy. Overall it achieves a good balance between accuracy and computation.