Single Shot Multibox Detector Research Articles

Aura detection using thermal camera with convolutional neural network method for mental health diagnosis

Mental health is an important aspect in realizing overall health. For this reason, non-invasive medical equipment is needed for people with mental disorders. This study aimed to create a psychological health diagnostic tool by detecting auras using a thermal camera from facial objects. The contribution of this study is that the tool can detect the patient's aura without physical contact so that the patient is more comfortable and does not feel invaded. This research designed a system for detecting electromagnetic wave radiation energy emitted by the body using a thermal camera. Face detection in the input image was performed using a convolutional neural network (CNN) model single shot multibox detector (SSD), which is one of the CNN models that implements a bounding box to estimate the localization of detected objects. In this case, system testing was used to evaluate the performance of the CNN system algorithm for aura detection in terms of color (main color or average color). The results obtained were detectibility by 80%, selectivity by 88.88%, precision by 70%, sensitivity by 87.5%, and accuracy by 63.63%. The design of the aura detection system in this study will make it easier for psychiatrists and psychologists to help make a noninvasive diagnosis.

Image-Processing-Based Subway Tunnel Crack Detection System.

With the increase in urban rail transit construction, instances of tunnel disease are on the rise, and cracks have become the focus of tunnel maintenance and management. Therefore, it is essential to carry out crack detection in a timely and efficient manner to not only prolong the service life of the tunnel but also reduce the incidence of accidents. In this paper, the design and structure of a tunnel crack detection system are analyzed. On this basis, this paper proposes a new method for crack identification and feature detection using image processing technology. This method fully considers the characteristics of tunnel images and the combination of these characteristics with deep learning, while a deep convolutional network (Single-Shot MultiBox Detector (SSD)) is proposed based on deep learning for object detection in complex images. The experimental results show that the test set accuracy and training set accuracy of the support vector machine (SVM) in the classification comparison test are up to 88% and 87.8%, respectively; while the test accuracy of Alexnet's deep convolutional neural network-based classification and identification is up to 96.7%, and the training set accuracy is up to 97.5%. It can be seen that this deep convolutional network recognition algorithm based on deep learning and image processing is better and more suitable for the detection of cracks in subway tunnels.

Cognitive cloud framework for waste dumping analysis using deep learning vision computing in healthy environment

Increased population and usage of more unnatural particles create dangerous waste in society. Environmental Sustainability is ensured based on proper waste management solutions. There is a need for real-time monitoring of waste dumping in the respective area to alert the individual management team to control it. We specifically investigated a solution to the issue of preventing the illegal disposal of waste in smart cities. In this regard, we will discuss the working model of an automatic visual identification and warning system. The proposed method uses cognitive computing technologies to analyze films taken by cameras located in metropolitan areas. To locate huge trash waste where it shouldn't be and alert the municipality to the situation. Initially, the proposed method develops the application prototype. Next, the Deep learning method Single Shot MultiBox Detector (SSD) MobileNet v2 is used to detect bulky waste data. Finally, the CNN method classifies the data and triggers the alarm system. The alarm system alerts the municipality with warning sounds to remove waste dumping. The proposed method provides an intelligent waste management system and a healthy environment. The simulation outcomes demonstrate that the suggested strategy yields a superior outcome.

Transfer Learning-Based Lightweight SSD Model for Detection of Pests in Citrus

In citrus cultivation, it is a difficult task for farmers to classify different pests correctly and make proper decisions to prevent citrus damage. This work proposes an efficient modified lightweight transfer learning model which combines the effectiveness and accuracy of citrus pest characterization with mobile terminal counting. Firstly, we utilized typical transfer learning feature extraction networks such as ResNet50, InceptionV3, VGG16, and MobileNetV3, and pre-trained the single-shot multibox detector (SSD) network to compare and analyze the classification accuracy and efficiency of each model. Then, to further reduce the amount of calculations needed, we miniaturized the prediction convolution kernel at the end of the model and added a residual block of a 1 × 1 convolution kernel to predict category scores and frame offsets. Finally, we transplanted the preferred lightweight SSD model into the mobile terminals developed by us to verify its usability. Compared to other transfer learning models, the modified MobileNetV3+RPBM can enable the SSD network to achieve accurate detection of Panonychus Citri Mcgregor and Aphids, with a mean average precision (mAP) up to 86.10% and the counting accuracy reaching 91.0% and 89.0%, respectively. In terms of speed, the mean latency of MobileNetV3+RPBM is as low as 185 ms. It was concluded that this novel and efficient modified MobileNetV3+RPBM+SSD model is effective at classifying citrus pests, and can be integrated into devices that are embedded for mobile rapid detection as well as for counting pests in citrus orchards. The work presented herein can help encourage farm managers to judge the degree of pest damage and make correct decisions regarding pesticide application in orchard management.

Autonomous face mask detection using single shot multibox detector, and ResNet-50 with identity retrieval through face matching using deep siamese neural network.

The COVID-19 pandemic poses a global health challenge. The World Health Organization states that face masks are proven to be effective, especially in public areas. Real-time monitoring of face masks is challenging and exhaustive for humans. To reduce human effort and to provide an enforcement mechanism, an autonomous system has been proposed to detect non-masked people and retrieve their identity using computer vision. The proposed method introduces a novel and efficient method that involves fine-tuning the pre-trained ResNet-50 model with a new head layer for classification between masked and non-masked people. The classifier is trained using adaptive momentum optimization algorithm with decaying learning rate and binary cross-entropy loss. Data augmentation and dropout regularization are employed to achieve best convergence. During real-time application of our classifier on videos, a Caffe face detector model based on Single Shot MultiBox Detector is used to extract the face regions of interest from each frame, on which the trained classifier is applied for detecting the non-masked people. The faces of these people are then captured, which is passed on to a deep siamese neural network, based on VGG-Face model for face matching. The captured faces are compared with the reference images from the database, by extracting the features and calculating cosine distance. If the faces match, the details of that person are retrieved from the database and displayed on the web application. The proposed method has secured best results where the trained classifier has achieved 99.74% accuracy, and the identity retrieval model achieved 98.24% accuracy.

Deep Learning-Based Automatic Defect Detection Method for Sewer Pipelines

To address the issues of low automation, reliance on manual screening by professionals, and long detection cycles in current urban drainage pipeline defect detection, this study proposes an improved object detection algorithm called EFE-SSD (enhanced feature extraction SSD), based on the SSD (Single Shot MultiBox Detector) network. Firstly, the RFB_s module is added to the SSD backbone network to enhance its feature extraction capabilities. Additionally, multiple scale features are fused to improve the detection performance of small target defects to some extent. Then, the improved ECA attention mechanism is used to adjust the channel weights of the output layer, suppressing irrelevant features. Finally, the Focal Loss is employed to replace the cross-entropy loss in the SSD network, effectively addressing the issue of imbalanced positive and negative samples during training. This increases the weight of difficult-to-classify samples during network training, further improving the detection accuracy of the network. Experimental results show that EFE-SSD achieves a detection mAP of 92.2% for four types of pipeline defects: Settled deposits, Displaced joints, Deformations, and Roots. Compared to the SSD algorithm, the model’s mAP was increased by 2.26 percentage points—ensuring the accuracy of pipeline defect detection.

DSM-IDM-YOLO: Depth-wise Separable Module and Inception Depth-wise Module Based YOLO for Pedestrian Detection

Pedestrian detection is one of the most challenging research areas in computer vision. Compared to traditional hand-crafted methods, convolutional neural networks (CNNs) have superior detection results. The single-stage detection networks, particularly the state-of-the-art You Only Look Once (YOLO) network, have attained a satisfactory performance without compromising the computation speed in object detection. YOLO framework can be leveraged in pedestrian detection as well. In this work, we propose an improved YOLOv2, called DSM-IDM-YOLO. The proposed model uses a modified DarkNet19 integrated with three new modules, two depth-wise separable convolution modules and one inception depth-wise convolution module, leading to a comprehensive feature of an object in the image. The modules are integrated on top of features from multiple levels of the network. The proposed framework is computationally less expensive owing to its convolution design and a moderate number of layers. It aims to improve performance with minimal computational overhead. The proposed method is compared with state-of-the-art detection methods, i.e., Faster R-CNN, YOLOv2, YOLOv3, YOLOv4-tiny and Single Shot Multibox Detector (SSD). The performance results attest that the proposed method has effectively improved the detection. Three benchmark pedestrian datasets are used for experimental analysis: INRIA, PASCAL VOC 2012 and Caltech.

Image and video-based crime prediction using object detection and deep learning

In recent years, the use of artificial intelligence (AI) for image and video-based crime detection has gained significant attention from law enforcement agencies and security experts. Indeed, deep learning (DL) models can learn complex patterns from data and help law enforcement agencies save time and resources by automatically identifying and tracking potential criminals. This contributes to make deep investigations and better steer their targets’ searches. Among others, handheld firearms and bladed weapons are the most frequent objects encountered at crime scenes. In this paper, we propose a DL-based surveillance system that can detect the presence of tracked objects, such as handheld firearms and bladed weapons, as well as may proceed to alert authorities regarding eventual threats before an incident occurs. After making a comparison of different DL-based object detection techniques, such as you only look once (YOLO), single shot multibox detector (SSD), or faster region-based convolutional neural networks (R-CNN), YOLO achieves the optimal balance of mean average precision (mAP) and inference speed for real-time prediction. Thus, we retain YOLOv5 for the implementation of our solution.

Deep Learning-Based Cow Tail Detection and Tracking for Precision Livestock Farming

Cow tail detection and tracking in videos provide valuable information for individual identification, calving process, behavior analysis, and body condition monitoring. Although deep learning-based detection methods have demonstrated good performance, many of them have high complexity and still require an improvement in video object detection by reducing the computation time and false positives. To fully exploit the interframe information and achieve a real-time online detection, the optimized cow tail detection and tracking method is proposed based on an improved single shot multibox detector (SSD) and Kalman filter. Here, our improved SSD-integrated DenseNet and Inception-v4 reduce detection information loss and network parameters. Then, an improved window function-based Kalman filter and Hungarian are adopted to remove error detections and enhance the cow tail tracking accuracy. Experiments on our acquired rear-view videos show that the proposed approach achieved fast tail detection with an accuracy of 96.97%, a speed of 96 fps, a smaller model size of 25 MB, and higher position accuracy with an average 6.45 pixels deviation. The proposed approach outperformed the region-based R-CNN models and other tracking methods (e.g., particle filter), which provides a new solution to automatic cow detection and tracking in smart livestock farming.

Analysis of object recognition trends based on deep learning

With the increasing development and maturity of deep learning, computers have also made world-renowned achievements in the domain of vision, especially in the basic and core branch of object detection, giving birth to many classical algorithms, which are widely used in many fields such as autonomous driving, intelligent medical care, intelligent security, and search entertainment. Before the emergence of deep learning algorithms, traditional algorithms for object detection were usually divided into three stages: region selection, feature extraction, and feature classification. However, with the advent of deep learning algorithms, object detection has taken to another peak, with Single Shot MultiBox Detector (SSD) enabling first-order detection of multi-feature maps and Region-based Convolutional Neural Networks (R-CNN) improving the performance of object detection while enabling instance segmentation. For object detection, this paper investigates the traditional algorithms, R-CNN, SSD, You Only Look Once (YOLO), and diffusion model, which is influential detection algorithms, and compares their differences as well as advantages in object detection to provide a reference for related research.

Improving the Performance of the Single Shot Multibox Detector for Steel Surface Defects with Context Fusion and Feature Refinement

Strip surface defects have large intraclass and small interclass differences, resulting in the available detection techniques having either a low accuracy or very poor real-time performance. In order to improve the ability for capturing steel surface defects, the context fusion structure introduces the local information of the shallow layer and the semantic information of the deep layer into multiscale feature maps. In addition, for filtering the semantic conflicts and redundancies arising from context fusion, a feature refinement module is introduced in our method, which further improves the detection accuracy. Our experimental results show that this significantly improved the performance. In particular, our method achieved 79.5% mAP and 71 FPS on the public NEU-DET dataset. This means that our method had a higher detection accuracy compared to other techniques.

Improving video surveillance systems in banks using deep learning techniques

In the contemporary world, security and safety are significant concerns for any country that wants to succeed in tourism, attracting investors, and economics. Manually, guards monitoring 24/7 for robberies or crimes becomes an exhaustive task, and real-time response is essential and helpful for preventing armed robberies at banks, casinos, houses, and ATMs. This paper presents a study based on real-time object detection systems for weapons auto-detection in video surveillance systems. We propose an early weapon detection framework using state-of-the-art, real-time object detection systems such as YOLO and SSD (Single Shot Multi-Box Detector). In addition, we considered closely reducing the number of false alarms in order to employ the model in real-life applications. The model is suitable for indoor surveillance cameras in banks, supermarkets, malls, gas stations, and so forth. The model can be employed as a precautionary system to prevent robberies by implying the model in outdoor surveillance cameras.

Implementasi Sistem Penghitung Kendaraan Otomatis Berbasis Computer Vision

The development of computer technology today is very helpful for humans in completing their work in various fields. One application of computer technology i.e., in the field of computer vision which has a very important role for object recognition. In this study, we designed a computer vision-based automatic vehicle counting system. The system that we created uses the MobileNetV2 Single Shot Multibox Detector (SSD) which is placed on the Raspberry Pi 4 to carry out the process of classifying cars and motorcycles and the raspberry pi 4 also functions as a system controller. This automatic vehicle counter system has been integrated between Raspberry Pi 4 and a mobile application on a smartphone where the smartphone functions to display information such as day, date, month, year and together with the number of cars and motorcycles. We tested this automatic vehicle counting system on steam services (car and motorcycle washing) for 3 days where 10 vehicles were collected every day. The test results show that the system is capable of detecting cars and motorcyles with an average accuracy rate of 46.6%.

OBJECT DETECTION IN AN IMAGE

This project aims to develop an object detection system using TensorFlow.js, a JavaScript library for building and training machine learning models in the browser or Node.js environment. Object detection is a critical task in computer vision, with various applications such as image and video analysis, robotics, and autonomous driving. The proposed system uses a pre-trained deep learning model such as MobileNet or ResNet as the feature extractor, and applies the Single Shot MultiBox Detector (SSD) algorithm for object detection. The system is trained and fine-tuned on a custom dataset, and the resulting model is integrated with a web-based user interface for real-time object detection. The performance of the system is evaluated on a test set and compared with existing object detection models

High accuracy intelligent real-time framework for detecting infant drowning based on deep learning

Timely detection of swimming accidents plays a critical role in keeping the safety of infants in pools. However, the targets in the infant drowning detection are generally small, densely-distributed and often immersed in complicated environment with lighting various and toy disturbances, and thus there are currently few efficient, accurate and comprehensive solutions for such a task. Aiming at this problem, a novel intelligent real-time framework for detecting infant drowning is proposed. Processing of the framework includes two stages: in stage I, an attention mechanism-integrated YOLOv5-alike model is designed to extract infant portrait foreground, which can reduce the interferences caused by background noise and improve the accuracy of detecting tiny objects. On this basis, in stage II, Single Shot Multi-box Detector (SSD) is utilized to filter out the falsely-detected targets in stage I and recognize swimming posture accurately. Live videos are collected from ten infant swimming pools to generate training dataset of totally 7,296 swimming images and 7,723 portrait images of infants for stage I and II respectively. And a set of 1,222 infant swimming images from another two pools are annotated to test the trained framework. Experimental results show that the Mean Average Precision (mAP) of the framework is 97.17%, and the processing speed can reach 43 frames per second, which outperforms any previous related network and has significant application value in practical infant drowning detection. Additionally, a web-based platform is developed to put the framework into practice, whose preliminary test results demonstrate its strong application potential.

Diagnostic accuracy of a deep learning model using YOLOv5 for detecting developmental dysplasia of the hip on radiography images

Developmental dysplasia of the hip (DDH) is a cluster of hip development disorders and one of the most common hip diseases in infants. Hip radiography is a convenient diagnostic tool for DDH, but its diagnostic accuracy is dependent on the interpreter’s level of experience. The aim of this study was to develop a deep learning model for detecting DDH. Patients younger than 12 months who underwent hip radiography between June 2009 and November 2021 were selected. Using their radiography images, transfer learning was performed to develop a deep learning model using the “You Only Look Once” v5 (YOLOv5) and single shot multi-box detector (SSD). A total of 305 anteroposterior hip radiography images (205 normal and 100 DDH hip images) were collected. Of these, 30 normal and 17 DDH hip images were used as the test dataset. The sensitivity and the specificity of our best YOLOv5 model (YOLOv5l) were 0.94 (95% confidence interval [CI] 0.73–1.00) and 0.96 (95% CI 0.89–0.99), respectively. This model also outperformed the SSD model. This is the first study to establish a model for detecting DDH using YOLOv5. Our deep learning model provides good diagnostic performance for DDH. We believe our model is a useful diagnostic assistant tool.

Tree Recognition and Crown Width Extraction Based on Novel Faster-RCNN in a Dense Loblolly Pine Environment

Tree crown width relates directly to wood quality and tree growth. The traditional method used to measure crown width is labor-intensive and time-consuming. Pairing imagery taken by an unmanned aerial vehicle (UAV) with a deep learning algorithm such as a faster region-based convolutional neural network (Faster-RCNN) has the potential to be an alternative to the traditional method. In this study, Faster-RCNN outperformed single-shot multibox detector (SSD) for crown detection in a young loblolly pine stand but performed poorly in a dense, mature loblolly pine stand. This paper proposes a novel Faster-RCNN algorithm for tree crown identification and crown width extraction in a forest stand environment with high-density loblolly pine forests. The new algorithm uses Residual Network 101 (ResNet101) and a feature pyramid network (FPN) to build an FPN_ResNet101 structure, improving the capability to model shallow location feature extraction. The algorithm was applied to images from a mature loblolly pine plot in eastern Texas, USA. The results show that the accuracy of crown recognition and crown width measurement using the FPN_ResNet101 structure as the backbone network in Faster-RCNN (FPN_Faster-RCNN_ResNet101) was high, being 95.26% and 0.95, respectively, which was 4.90% and 0.27 higher than when using Faster-RCNN with ResNet101 as the backbone network (Faster-RCNN_ResNet101). The results fully confirm the effectiveness of the proposed algorithm.

Research on a Target Detection Method for Alternanthera Philoxeroides in the Rice Seedling Stage Based on Single-Shot Multibox Detector

Alternanthera philoxeroides, an invasive alien malignant weed, competes with rice for water, fertilizer, light, and growth space before seedling closure stages, which commonly stresses the growth of rice. Chemical herbicides are mainly used to control weeds. However, excessive use of chemical herbicides could lead to serious environmental pollution. With the rapid development of artificial intelligence and deep-learning techniques, it is possible to reduce the use of chemical herbicides by spraying herbicides on only precise regions of weeds in rice fields. To improve the accuracy of the model in identifying weed regions, in this study the performance among the you only look once (YOLO) series and single-shot multibox detector (SSD) one-stage target detection models, i.e., YOLOv3, YOLOv4-tiny, YOLOv5-s, and SSD target detection networks, is discussed. The experimental results showed that the SSD-based target detection model for Alternanthera philoxeroides was better than the YOLO series due to its higher recall, mAP (mean average precision), and F1 values, which reached 0.874, 0.942, and 0.881, respectively. Meanwhile, the target detection model based on SSD performed better than the YOLO series when dealing with mutual occlusion images between seedlings and Alternanthera philoxeroides. In conclusion, in this study a high-accuracy method is provided for detecting precise regions of Alternanthera philoxeroides by constructing a model based on SSD, contributing to the reduction of environmental pollution.

Identification method of typical defects in transmission lines based on YOLOv5 object detection algorithm

In order to realize the automatic identification of typical defects in transmission lines, a typical defect identification method for transmission lines based on You Only Look Once (YOLO)v5 algorithm is proposed, which has the advantages of low missing detection rate and false detection rate. Firstly, a transmission line defect image dataset is constructed for three typical defects: bird’s nest defect, insulator defect and shakeproof hammer defect, and the corresponding image labeling strategy is formulated according to the defect characteristics. Secondly, this paper introduces the network structure of YOLOv5 in detail. For the sake of further improving the neural network’s ability to recognize the visual defects of transmission lines, the attention mechanism and small target detection layer are introduced into YOLOv5 neural network. Finally, this paper design transmission line defect recognition cases based on three kinds of typical defects, and the proposed method is compared with the recognition methods based on Faster Region based Convolutional Neural Network (RCNN) and Single Shot MultiBox Detector (SSD) in terms of the proposed performance indexes (missing detection rate and false detection rate). The results show that the missing detection rate and false detection rate of the YOLOv5-based identification method are much lower than those of the Faster RCNN and SSD-based identification methods, and the improved YOLOv5 network recognition rate is higher, which is significant for reducing the workload of transmission line inspection personnel.

Single shot multibox detector object detection based on attention mechanism and feature fusion

The single shot multibox detector (SSD) is one of the most important algorithms in single-stage target detection, having the characteristics of multiscale detection and rapid detection speed. However, the effective SSD feature layers are independent of one another, which can lead to object detection difficulties. To address this problem, we proposed an improved SSD object detection algorithm. First, the global attention mechanism (GAM)—which can enhance spatial and channel information—was introduced into the multiscale feature layer. The channel attention module of the GAM was improved. Second, a feature fusion module was introduced to strengthen the relationship between feature layers. Finally, the cross stage partial structure was introduced into the feature fusion module, and used to improve the model’s learning ability. For model training and detection based on the PASCAL VOC dataset, the mean average precision and frames per second obtained by the improved SSD algorithm were 84.67% and 18.67, respectively, which could effectively detect difficult targets.