Region-based Convolutional Network Research Articles

Comparison of faster region-based convolutional network for algorithms for grape leaves classification

Comparison of faster region-based convolutional network for algorithms for grape leaves classification

Comparison and Optimization of Rail Defect Detection Methods Based on Object Detection Model

Rail maintenance is an extremely complex but important component of a metro system. In response to the frequent occurrence of rail defects, the rail defect detection based on object detection is compared and optimized. First, the detection performance of the Single Shot MultiBox Detector (SSD) model and the Fast Region-based Convolutional Network (Faster R-CNN) model on rail defect are compared through model training and validation. Second, to address the issue of low defect visibility in images, the rail defect detection method combining the grayscale histogram equalization and the Faster R-CNN model is proposed. Finally, to address the issue of failed detections in rail defect detection, the Faster R-CNN model is improved by adding the Feature Pyramid Network (FPN) structure. Results show that the rail defect detection method using the Faster R-CNN model is more accurate, with the rail defects of mean average precision (mAP) is 75%. By performing grayscale histogram equalization, resulting in a 2-percentage point increase in the mAP0.75 of the rail defect detection based on the Faster R-CNN model. By combining the FPN with Faster R-CNN model, the mAP of small-scale defect detection is increased by 15 percentage points, and the AP of various rail defects is increased by 3, 13, 13, and 15 percentage points, respectively.

Performance evaluation of semi-supervised learning frameworks for multi-class weed detection.

Precision weed management (PWM), driven by machine vision and deep learning (DL) advancements, not only enhances agricultural product quality and optimizes crop yield but also provides a sustainable alternative to herbicide use. However, existing DL-based algorithms on weed detection are mainly developed based on supervised learning approaches, typically demanding large-scale datasets with manual-labeled annotations, which can be time-consuming and labor-intensive. As such, label-efficient learning methods, especially semi-supervised learning, have gained increased attention in the broader domain of computer vision and have demonstrated promising performance. These methods aim to utilize a small number of labeled data samples along with a great number of unlabeled samples to develop high-performing models comparable to the supervised learning counterpart trained on a large amount of labeled data samples. In this study, we assess the effectiveness of a semi-supervised learning framework for multi-class weed detection, employing two well-known object detection frameworks, namely FCOS (Fully Convolutional One-Stage Object Detection) and Faster-RCNN (Faster Region-based Convolutional Networks). Specifically, we evaluate a generalized student-teacher framework with an improved pseudo-label generation module to produce reliable pseudo-labels for the unlabeled data. To enhance generalization, an ensemble student network is employed to facilitate the training process. Experimental results show that the proposed approach is able to achieve approximately 76% and 96% detection accuracy as the supervised methods with only 10% of labeled data in CottonWeedDet3 and CottonWeedDet12, respectively. We offer access to the source code (https://github.com/JiajiaLi04/SemiWeeds), contributing a valuable resource for ongoing semi-supervised learning research in weed detection and beyond.

Black-box adversarial patch attacks using differential evolution against aerial imagery object detectors

Adversarial patch attacks are commonly used to attack aerial imagery object detectors. However, most existing methods are designed for white-box settings, which are impractical in real world scenarios. In this work, we propose a novel framework for black-box adversarial patch attacks against aerial imagery object detectors using differential evolution (DE). Specifically, we first introduce a new dimensionality reduction strategy to address the high-dimensionality curse in DE optimization problems. Then, we design three universal fitness functions to help DE to select better individuals within a limited computational budget. Finally, we conduct extensive experiments on the Dataset for Object Detection in Aerial images (DOTA) against You Only Look Once (YOLOv3, YOLOv4), and Faster Region-based Convolutional Network (Faster R-CNN), utilizing attack success rate (ASR) and average precision (AP) to quantitatively evaluate the attack efficiency. Results on YOLOv3 for the plane category indicate that our proposed strategy achieves at least a 20.97% improvement in ASR compared to existing dimensionality reduction schemes. Moreover, experiments conducted on detectors with different architectures reveal inconsistent adversarial robustness across categories. To the best of our knowledge, this is the first work to explore the use of DE in black-box adversarial patch attacks against aerial imagery object detectors. Our code has been released at https://github.com/tang-agui/Bbox-Att-Detector.

Comparative Evaluation of Convolutional Neural Network Object Detection Algorithms for Vehicle Detection.

The domain of object detection was revolutionized with the introduction of Convolutional Neural Networks (CNNs) in the field of computer vision. This article aims to explore the architectural intricacies, methodological differences, and performance characteristics of three CNN-based object detection algorithms, namely Faster Region-Based Convolutional Network (R-CNN), You Only Look Once v3 (YOLO), and Single Shot MultiBox Detector (SSD) in the specific domain application of vehicle detection. The findings of this study indicate that the SSD object detection algorithm outperforms the other approaches in terms of both performance and processing speeds. The Faster R-CNN approach detected objects in images with an average speed of 5.1 s, achieving a mean average precision of 0.76 and an average loss of 0.467. YOLO v3 detected objects with an average speed of 1.16 s, achieving a mean average precision of 0.81 with an average loss of 1.183. In contrast, SSD detected objects with an average speed of 0.5 s, exhibiting the highest mean average precision of 0.92 despite having a higher average loss of 2.625. Notably, all three object detectors achieved an accuracy exceeding 99%.

Leveraging optimal UAV-lidar features for two CNN approaches to delineate mangrove tree crowns

Effective individual tree crown delineation plays a critical role for assessing mangrove quality and health. However, mangrove tree crowns often form large, interconnected clusters in dense coverage areas, making it difficult to separate them individually. Moreover, many mangrove trees have multiple large branches that result in irregular tree shapes and create internal gaps inside their crowns. Unmanned Aerial Vehicle (UAV) lidar data can potentially overcome these challenges, but a comprehensive assessment of UAV-lidar-derived features is still missing, as well as how to best use them for dense mangrove forests. In this study, we set forth two objectives: (1) to derive optimal features for mangrove individual tree crown delineation through an exclusive examination of all the UAV-lidar features; (2) to develop effective methods that can best incorporate these optimal UAV-lidar features. To achieve the first objective, we extracted 224 features from UAV-lidar data based on three groups of attributes: height, intensity, and point density. Seven spatial scales ranging from 0.2 to 0.5 m were tested in this process. For the second objective, we applied two state-of-the-art Convolutional Neural Networks: Mask Region-based Convolutional Network method (Mask R–CNN) and Ultralytics You Only Look Once version 8 (YOLOv8). At last, the derived three optimal features are: canopy height model at 0.20 m, coefficient of point cloud height variation at 0.25 m, and ground point percentage at 0.25 m. Comparing our methods with traditional methods that only use canopy height model, we found that integrating Convolutional Neural Networks and the optimal UAV-lidar features improved the accuracy by more than 13%. Mask R–CNN was better for dense mangroves, while YOLOv8 was excellent for sparse and short mangroves. The derived optimal UAV-lidar features further enhanced the detection of short trees, densely clumped trees and trees with irregular shapes. To conclude, we developed three novel features based on UAV-lidar data that are especially suitable for individual tree crown delineation in dense mangrove forests. Using these features, we demonstrated that Convolutional Neural Networks can achieve high performance. We hope that our methods will facilitate various mangrove forest studies that rely on accurate individual tree crown delineation.

Head-Integrated Detecting Method for Workers under Complex Construction Scenarios

Real-time detection of workers is crucial in construction safety management. Deep learning-based detecting methods are valuable, but always challenged by the possibility of target missing or identity errors under complex scenarios. To address these limitations, previous research depended on re-training for new models or datasets, which are prohibitively time-consuming and incur high computing demands. However, we demonstrate that the better detecting model might not rely on more re-training of weights; instead, a training-free model can achieve even better performance by integrating head information. In this paper, a new head-detecting branch (55 MB) is added to the Keypoint Region-based Convolutional Network (Keypoint R-CNN, 226 MB) without altering its original weights, allowing for a less occluded head to aid in body detection. We also deployed motion information and anthropometric data through a post-processing module to calculate movement relationships. This study achieved an identity F1-score (IDF1) of 97.609%, recall (Rcll) of 98.173%, precision (Prcn) of 97.052%, and accuracy of 95.329% as a state-of-the-art (SOTA) method for worker detection. This exploration breaks the inertial attitudes of re-training dependency and accelerates the application of universal models, in addition to reducing the computational difficulty for most construction sites, especially in scenarios with an insufficient graphics processing unit (GPU). More importantly, this study can address occlusion challenges effectively in the worker detection field, making it of practical significance.

An Automated Instance Segmentation Method for Crack Detection Integrated with CrackMover Data Augmentation.

Crack detection plays a critical role in ensuring road safety and maintenance. Traditional, manual, and semi-automatic detection methods have proven inefficient. Nowadays, the emergence of deep learning techniques has opened up new possibilities for automatic crack detection. However, there are few methods with both localization and segmentation abilities, and most perform poorly. The consistent nature of pavement over a small mileage range gives us the opportunity to make improvements. A novel data-augmentation strategy called CrackMover, specifically tailored for crack detection methods, is proposed. Experiments demonstrate the effectiveness of CrackMover for various methods. Moreover, this paper presents a new instance segmentation method for crack detection. It adopts a redesigned backbone network and incorporates a cascade structure for the region-based convolutional network (R-CNN) part. The experimental evaluation showcases significant performance improvements achieved by these approaches in crack detection. The proposed method achieves an average precision of 33.3%, surpassing Mask R-CNN with a Residual Network 50 backbone by 8.6%, proving its effectiveness in detecting crack distress.

Lightweight Detection Methods for Insulator Self-Explosion Defects.

The accurate and efficient detection of defective insulators is an essential prerequisite for ensuring the safety of the power grid in the new generation of intelligent electrical system inspections. Currently, traditional object detection algorithms for detecting defective insulators in images face issues such as excessive parameter size, low accuracy, and slow detection speed. To address the aforementioned issues, this article proposes an insulator defect detection model based on the lightweight Faster R-CNN (Faster Region-based Convolutional Network) model (Faster R-CNN-tiny). First, the Faster R-CNN model's backbone network is turned into a lightweight version of it by substituting EfficientNet for ResNet (Residual Network), greatly decreasing the model parameters while increasing its detection accuracy. The second step is to employ a feature pyramid to build feature maps with various resolutions for feature fusion, which enables the detection of objects at various scales. In addition, replacing ordinary convolutions in the network model with more efficient depth-wise separable convolutions increases detection speed while slightly reducing network detection accuracy. Transfer learning is introduced, and a training method involving freezing and unfreezing the model is employed to enhance the network's ability to detect small target defects. The proposed model is validated using the insulator self-exploding defect dataset. The experimental results show that Faster R-CNN-tiny significantly outperforms the Faster R-CNN (ResNet) model in terms of mean average precision (mAP), frames per second (FPS), and number of parameters.

Raising the curtain: Bubble size measurement inside parallel plate electrolyzers

Alongside vertical gas-evolving electrodes, gas bubbles nucleate, grow, coalesce, detach, and ultimately form a bubble curtain, affecting the fluid dynamic, electrochemical reaction, mass transport, and ohmic drop. Due to the small inter-electrode gap in parallel plate electrolyzers, the size determination of detached bubbles inside the bubble curtain close to the electrodes is challenging. In particular, the size determination of detached oxygen bubbles needs to be addressed. We present an optical setup to record images from detached oxygen bubbles inside the bubble curtain of a membrane-separated, parallel plate electrolyzer. The images are postprocessed using an adopted region-based convolutional neuronal network to quantify the bubble sizes. Next, the electrolyte flow velocity, concentration, current density, and catalysts are varied, and the effect on the oxygen bubble size distribution is studied. The median size ranges from 140 to ▪, doubling values from literature measured in artificial electrolyzers.

Bi-LS-AttM: A Bidirectional LSTM and Attention Mechanism Model for Improving Image Captioning

The discipline of automatic image captioning represents an integration of two pivotal branches of artificial intelligence, namely computer vision (CV) and natural language processing (NLP). The principal functionality of this technology lies in transmuting the extracted visual features into semantic information of a higher order. The bidirectional long short-term memory (Bi-LSTM) has garnered wide acceptance in executing image captioning tasks. Of late, scholarly attention has been focused on modifying suitable models for innovative and precise subtitle captions, although tuning the parameters of the model does not invariably yield optimal outcomes. Given this, the current research proposes a model that effectively employs the bidirectional LSTM and attention mechanism (Bi-LS-AttM) for image captioning endeavors. This model exploits the contextual comprehension from both anterior and posterior aspects of the input data, synergistically with the attention mechanism, thereby augmenting the precision of visual language interpretation. The distinctiveness of this research is embodied in its incorporation of Bi-LSTM and the attention mechanism to engender sentences that are both structurally innovative and accurately reflective of the image content. To enhance temporal efficiency and accuracy, this study substitutes convolutional neural networks (CNNs) with fast region-based convolutional networks (Fast RCNNs). Additionally, it refines the process of generation and evaluation of common space, thus fostering improved efficiency. Our model was tested for its performance on Flickr30k and MSCOCO datasets (80 object categories). Comparative analyses of performance metrics reveal that our model, leveraging the Bi-LS-AttM, surpasses unidirectional and Bi-LSTM models. When applied to caption generation and image-sentence retrieval tasks, our model manifests time economies of approximately 36.5% and 26.3% vis-a-vis the Bi-LSTM model and the deep Bi-LSTM model, respectively.

A quantitative heterogeneity analysis approach to molecular subtype recognition of breast cancer in dynamic contrast-enhanced magnetic imaging images from radiomics data.

Breast cancer is a major cause of mortality among women worldwide. Dynamic contrast-enhanced breast magnetic resonance imaging (DCE-MRI) is a good imaging technique that can show temporal information about the kinetics of the contrast agent in suspicious breast lesions as well as acceptable spatial resolution. Computer-aided detection systems assist in the detection of lesions through medical image processing techniques combined with computerized analysis and calculation, which in turn helps radiologists recognize molecular subtypes of breast lesions that will be beneficial for better treatment plan decisions. In this paper, a computer-aided diagnosis method is proposed to automatically locate breast cancer lesions and identify molecular subtypes of breast cancer with heterogeneity analysis from radiomics data. A fast region-based convolutional network (Faster R-CNN) framework is first applied to images to detect breast cancer lesions. Then, the heterogeneous regions of every breast cancer lesion are extracted. Based on the multiple visual and kinetic radiomics features extracted from the heterogeneous regions, a temporal bag of visual word model is proposed, which takes into account the dynamic characteristics of both lesion and heterogeneous regions in images over time. The recognition task of molecular subtypes of breast lesions is realized based on a stacking classification model. At the genetic level, breast cancer is divided into four molecular subtypes, namely, luminal epithelial type A (Luminal A), luminal epithelial type B (Luminal B), HER-2 overexpression and basal cell type. The experimental results show that the precision of the four subtypes is 93%, 94%, 83%, 86%; the recall is 96%, 80%, 91%, 94%; and the F1-score is 95%, 86%, 87%. The experimental results denote the influence of heterogeneous regions on the recognition task. The DCE-MRI-based approach to identify molecular typing of breast cancer for noninvasive diagnosis will contribute to the development of breast cancer treatment, improved outcomes and reduced mortality.

LAD-RCNN: A Powerful Tool for Livestock Face Detection and Normalization.

With the demand for standardized large-scale livestock farming and the development of artificial intelligence technology, a lot of research in the area of animal face detection and face identification was conducted. However, there are no specialized studies on livestock face normalization, which may significantly reduce the performance of face identification. The keypoint detection technology, which has been widely applied in human face normalization, is not suitable for animal face normalization due to the arbitrary directions of animal face images captured from uncooperative animals. It is necessary to develop a livestock face normalization method that can handle arbitrary face directions. In this study, a lightweight angle detection and region-based convolutional network (LAD-RCNN) was developed, which contains a new rotation angle coding method that can detect the rotation angle and the location of the animal's face in one stage. LAD-RCNN also includes a series of image enhancement methods to improve its performance. LAD-RCNN has been evaluated on multiple datasets, including a goat dataset and infrared images of goats. Evaluation results show that the average precision of face detection was more than 97%, and the deviations between the detected rotation angle and the ground-truth rotation angle were less than 6.42° on all the test datasets. LAD-RCNN runs very fast and only takes 13.7 ms to process a picture on a single RTX 2080Ti GPU. This shows that LAD-RCNN has an excellent performance in livestock face recognition and direction detection, and therefore it is very suitable for livestock face detection and normalization.

AI-assisted identification of intrapapillary capillary loops in magnification endoscopy for diagnosing early-stage esophageal squamous cell carcinoma: a preliminary study.

Esophageal squamous cell carcinoma (ESCC) is one of the most common histological types of esophageal cancers. It can seriously affect public health, particularly in Eastern Asia. Early diagnosis and effective therapy of ESCC can significantly help improve patient prognoses. The visualization of intrapapillary capillary loops (IPCLs) under magnification endoscopy (ME) can greatly support the identification of ESCC occurrences by endoscopists. This paper proposes an artificial-intelligence-assisted endoscopic diagnosis approach using deep learning for localizing and identifying IPCLs to diagnose early-stage ESCC. An improved Faster region-based convolutional network (R-CNN) with a polarized self-attention (PSA)-HRNetV2p backbone was employed to automatically detect IPCLs in ME images. In our study, 2887 ME with blue laser imaging (ME-BLI) images of 246 patients and 493 ME with narrow-band imaging (ME-NBI) images of 81 patients were collected from multiple hospitals and used to train and test our detection model. The ME-NBI images were used as the external testing set to verify the generalizability of the model. The experimental evaluation revealed that the proposed method achieved a recall of 79.25%, precision of 75.54%, F1-score of 0.764 and mean average precision (mAP) of 74.95%. Our method outperformed other existing approaches in our evaluation. It can effectively improve the accuracy of ESCC detection and provide a useful adjunct to the assessment of early-stage ESCC for endoscopists.

Detection of Left Ventricular Cavity from Cardiac MRI Images Using Faster R-CNN

The automatic localization of the left ventricle (LV) in short-axis magnetic resonance (MR) images is a required step to process cardiac images using convolutional neural networks for the extraction of a region of interest (ROI). The precise extraction of the LV’s ROI from cardiac MRI images is crucial for detecting heart disorders via cardiac segmentation or registration. Nevertheless, this task appears to be intricate due to the diversities in the size and shape of the LV and the scattering of surrounding tissues across different slices. Thus, this study proposed a region-based convolutional network (Faster R-CNN) for the LV localization from short-axis cardiac MRI images using a region proposal network (RPN) integrated with deep feature classification and regression. The model was trained using images with corresponding bounding boxes (labels) around the LV, and various experiments were applied to select the appropriate layers and set the suitable hyper-parameters. The experimental findings show that the proposed model was adequate, with accuracy, precision, recall, and F1 score values of 0.91, 0.94, 0.95, and 0.95, respectively. This model also allows the cropping of the detected area of LV, which is vital in reducing the computational cost and time during segmentation and classification procedures. Therefore, it would be an ideal model and clinically applicable for diagnosing cardiac diseases.

Automatic image segmentation using Region-Based convolutional networks for Melanoma skin cancer detection

Melanoma is one of the most aggressive skin cancers, however, its early detection can significantly increase probabilities to cure it. Unfortunately, it is one of the most difficult skin cancers to detect, its detection relies mainly on the dermatologist’s expertise and experience with Melanoma. This research deals with targeting most of the common Melanoma stains or spots that could potentially evolve to Melanoma skin cancer. Region-based Convolutional Neural Networks were used as the model to detect and segment images of the skin area of interest. The neural network model is focused on providing instance segmentation rather than only a boxbounding object detection. The Mask R-CNN model was implemented to provide a solution for small trained datasets scenarios. Two pipelines were implemented, the first one was with only the Region-Based Convolutional Neural Network and the other one was a combined pipeline with a first stage using Mask R-CNN and then getting the result to use as feedback in a second stage implementing Grabcut, which is another segmentation method based on graphic cuts. Results demonstrated through Dice Similarity Coefficient and Jaccard Index that Mask R-CNN alone performed better in proper segmentation than Mask R-CNN + Grabcut model. In both models’ results, variation was very small when the training dataset size changed between 160, 100, and 50 images. In both of the pipelines, the models were capable of running the segmentation correctly, which illustrates that focalization of the zone is possible with very small datasets and the potential use of automatic segmentation to assist in Melanoma detection.

Detection of residual feed in aquaculture using YOLO and Mask RCNN

A dual-strategy residual feed detection algorithm based on improved You Only Look Once (YOLO) v3 and Mask Region-Based Convolutional Networks (RCNN) was explored for detecting residual feed in aquaculture areas. After improving the YOLO v3 network structure, the program can better extract the boundary box of swimming crabs and recognize tiny targets. Using the customized Mask RCNN, the region proposal network (RPN) traversed the fixed-size prior frame for small targets, which reduced false negatives in the inference process and speeded up the network. Using the YOLO v3 output, which extracts image information for swimming crab, as the input for Mask RCNN, the RPN recommended region and the probability of background misidentification were further reduced, which made the improved network more rapid and accurate. The precision of the local rapid detection model was 84%, the real-time detection time was 0.05 s, the detection precision was 96.5%, and the recall rate was 97%. Thus, the dual-strategy detection algorithm for residual feed developed herein could accurately identify residual feed and effectively reduce feed waste.

Copy move forgery detection and segmentation using improved mask region-based convolution network (RCNN)

Copy-move forgery (CMF) is a common image manipulation approach that uses the information from the same sample to manipulate it with the intent of hiding the required content. Several approaches have been designed for the timely detection of CMF; however, accurate identification of manipulated samples is a complicated job due to the similar capturing conditions of the copied content as the patch is taken from the same image. Moreover, the occurrence of several post-processing attacks i.e., noise, blurring, brightness variations, etc. further enhances the difficulties of the detection approaches. In this work, we attempted to cover the limitations of existing methods by proposing a deep learning (DL)-based approach for the accurate detection of CMF. A custom Mask-RCNN model with the DenseNet-41 as the base network is presented which is capable of nominating a better set of image features and presents the complex image transformation effectively. More descriptively, the DenseNet-41 model is used as the base network for deep keypoints extraction which is then localized, segmented, and categorized by the Mask-RCNN model to locate the manipulated area. We have tested the proposed model on three standard databases namely the CoMoFoD, MICC-F2000, and CASIA-v2 databases, and attained a precision of 98.12%, 99.02%, and 83.41%, respectively. We have reported the results for numerous image post-processing attacks and confirmed that the presented work is robust to detect the CMF in the presence of translation, scale variations, rotation, color changes, noise, compression, and blurring in images. We have confirmed through extensive quantitative and qualitative evaluation that the DenseNet-41-based Mask-RCNN model is robust to CMF detection and can assist forensic analyzers to detect forensic manipulations accurately.

Tooth-Related Disease Detection System Based on Panoramic Images and Optimization Through Automation: Development Study.

Early detection of tooth-related diseases in patients plays a key role in maintaining their dental health and preventing future complications. Since dentists are not overly attentive to tooth-related diseases that may be difficult to judge visually, many patients miss timely treatment. The 5 representative tooth-related diseases, that is, coronal caries or defect, proximal caries, cervical caries or abrasion, periapical radiolucency, and residual root can be detected on panoramic images. In this study, a web service was constructed for the detection of these diseases on panoramic images in real time, which helped shorten the treatment planning time and reduce the probability of misdiagnosis. This study designed a model to assess tooth-related diseases in panoramic images by using artificial intelligence in real time. This model can perform an auxiliary role in the diagnosis of tooth-related diseases by dentists and reduce the treatment planning time spent through telemedicine. For learning the 5 tooth-related diseases, 10,000 panoramic images were modeled: 4206 coronal caries or defects, 4478 proximal caries, 6920 cervical caries or abrasion, 8290 periapical radiolucencies, and 1446 residual roots. To learn the model, the fast region-based convolutional network (Fast R-CNN), residual neural network (ResNet), and inception models were used. Learning about the 5 tooth-related diseases completely did not provide accurate information on the diseases because of indistinct features present in the panoramic pictures. Therefore, 1 detection model was applied to each tooth-related disease, and the models for each of the diseases were integrated to increase accuracy. The Fast R-CNN model showed the highest accuracy, with an accuracy of over 90%, in diagnosing the 5 tooth-related diseases. Thus, Fast R-CNN was selected as the final judgment model as it facilitated the real-time diagnosis of dental diseases that are difficult to judge visually from radiographs and images, thereby assisting the dentists in their treatment plans. The Fast R-CNN model showed the highest accuracy in the real-time diagnosis of dental diseases and can therefore play an auxiliary role in shortening the treatment planning time after the dentists diagnose the tooth-related disease. In addition, by updating the captured panoramic images of patients on the web service developed in this study, we are looking forward to increasing the accuracy of diagnosing these 5 tooth-related diseases. The dental diagnosis system in this study takes 2 minutes for diagnosing 5 diseases in 1 panoramic image. Therefore, this system plays an effective role in setting a dental treatment schedule.

Pedestrian detection system based on deep learning

<p>Pedestrian detection is a rapidly growing field of computer vision with applications in smart cars, surveillance, automotive safety, and advanced robotics. Most of the success of the last few years has been driven by the rapid growth of deep learning, more efficient tools capable of learning semantic, high-level, deeper features of images are proposed. In this article, we investigated the task of pedestrian detection on roads using models based on convolutional neural networks. We compared the performance of standard state-of-the-art object detectors like Faster region-based convolutional network (R-CNN), single shot detector (SSD), and you only look once, version 3 (YOLOv3). Results show that YOLOv3 is the best object detection model than others for pedestrians in terms of detection and time prediction.</p>