scholarly journals Real-Time Human Detection in a Restricted Area for Safety in Truck Dumper Control System Using Deep Learning

2021 ◽  
pp. 29-35
Author(s):  
Apirak Worrakantapon ◽  
◽  
Wattana Pongsena ◽  
Kittisak Kerdprasop ◽  
Nittaya Kerdprasop
Keyword(s):  
Real Time ◽  
Network Architecture ◽  
High Speed ◽  
Animal Feed ◽  
Raw Materials ◽  
Human Detection ◽  
Control Room ◽  
Restricted Area ◽  
Detection Model ◽  
Safety Awareness

A process to receive raw materials from suppliers in an animal feed industry utilizes both automatic and semi-automatic machine control systems. The process called “truck dumper system” is the procedure that the suppliers provide raw materials carried by trucks; then, their tailgates open, and the raw materials are discharged by raising front end part of a truck to gather raw materials in a collection area. In general, the truck dumper system has been controlled manually by staff in a control room, not by a truck driver. However, serious accidents may occur during the process because when the dumper lifts up, the staff's vision has been blocked by the raised part of a truck. Therefore, if the staff controls the dumper to lift down by lacking safety awareness, people in the restricted area can be endangered. In this study, we proposed a framework of automatic human detection to prevent any accident that may occur from the truck dumper in the restricted area. The human detection model was developed to detect humans possibly in different blind corners that are difficult for staff in a control room to monitor these unseen areas for safety-awareness. The main technology of the proposed framework was the real-time human detection with fully convolutional neural network architecture called You Only Look Once, or YOLO. The framework has been designed to send a signal to terminate the truck dumper system immediately after the model detects people in the restricted area. In experiments, we discovered that the model could detect a human in all blind corners, including the corners that the staff's sight was completely bloacked by some barriers. The overall efficiency of this framework in an aspect of speed was high. The average time to process per image was 397 milliseconds by using CPUs and only 52 milliseconds by using GPUs. The results also showed that the model was effectively applicable to detect human in real-time due to its high-speed process.

Development of a Novel Convolutional Neural Network Architecture Named RoadweatherNet for Trajectory-Level Weather Detection using SHRP2 Naturalistic Driving Data

2021 ◽  
pp. 036119812110054
Author(s):  
Md Nasim Khan ◽  
Mohamed M. Ahmed
Keyword(s):  
Neural Network ◽  
Convolutional Neural Network ◽  
Real Time ◽  
Network Architecture ◽  
Video Data ◽  
Safe Driving ◽  
Detection Model ◽  
Weather Information ◽  
Adverse Weather ◽  
Naturalistic Driving

Driver performances could be significantly impaired in adverse weather because of poor visibility and slippery roadways. Therefore, providing drivers with accurate weather information in real time is vital for safe driving. The state-of-practice of collecting roadway weather information is based on weather stations, which are expensive and cannot provide trajectory-level weather information. Therefore, the primary objective of this study was to develop an affordable detection system capable of providing trajectory-level weather information at the road surface level in real-time. This study utilized the Strategic Highway Research Program 2 Naturalistic Driving Study video data combined with a promising machine learning technique, called convolutional neural network (CNN), to develop a weather detection model with seven weather categories: clear, light rain, heavy rain, light snow, heavy snow, distant fog, and near fog. A novel CNN architecture, named RoadweatherNet, was carefully crafted to achieve the weather detection task. The evaluation results based on a test dataset revealed that RoadweatherNet can provide excellent performance in detecting weather conditions with an overall accuracy of 93%. The performance of RoadweatherNet was also compared with six pre-trained CNN models, namely, AlexNet, ResNet18, ResNet50, GoogLeNet, ShuffleNet, and SqueezeNet, which showed that RoadweatherNet can provide nearly identical performance with a significant reduction in training time. The proposed weather detection model is cost-efficient and requires less computational power; therefore, it can be made widely available mainly owing to the recent thriving of smartphone cameras and can be used to expand and update the current weather-based variable speed limit systems.

A novel abnormal detection method for bearing temperature based on spatiotemporal fusion

2021 ◽  
pp. 095440972110221
Author(s):  
Yong Zhi Liu ◽  
Yi Sheng Zou ◽  
Yu Wu ◽  
Hao Yang Zhang ◽  
Guo Fu Ding
Keyword(s):  
Decision Making ◽  
Anomaly Detection ◽  
Real Time ◽  
Early Warning ◽  
High Speed ◽  
Entropy Method ◽  
High Speed Train ◽  
Detection Model ◽  
Time Prediction ◽  
Model Based

The existing bearing temperature fault detection and early warning system has a high false alarm rate and insufficient early warning ability. For this reason, in this study, a method for detecting the abnormal bearing temperature of high-speed trains based on spatiotemporal fusion decision-making was proposed. First, the temperature characteristics of similar bearings were compared and analyzed with different spatial distributions. Then, a bearing abnormal temperature rise detection model based on the analytic hierarchy process (AHP) entropy method was proposed. Second, the temperature characteristics of the same bearings were compared and analyzed with different time distributions. A real-time prediction model of high-speed train bearing temperature anomalies based on Bi-directional Long Short-Term Memory (BILSTM) was proposed. Finally, the D-S evidence theory was used to combine the anomaly detection model based on the AHP entropy method and the anomaly detection model based on BILSTM real-time prediction. Through the comprehensive diagnosis and decision-making of high-speed train bearings from two dimensions of space and time, a more comprehensive and accurate anomaly detection model was realized. The experimental results showed that the spatiotemporal comparison fusion decision model successfully eliminated the misjudgment phenomenon of single-dimension model diagnosis and that it has good early warning ability.

scholarly journals Depthwise Separable Convolution Neural Network for High-Speed SAR Ship Detection

2019 ◽  
Vol 11 (21) ◽  
pp. 2483 ◽  
Author(s):  
Zhang ◽  
Zhang ◽  
Shi ◽  
Wei
Keyword(s):  
Neural Network ◽  
Real Time ◽  
Network Architecture ◽  
High Speed ◽  
Fishery Management ◽  
Convolution Neural Network ◽  
Detection Accuracy ◽  
Ship Detection ◽  
Military Planning ◽  
Detection Speed

As an active microwave imaging sensor for the high-resolution earth observation, synthetic aperture radar (SAR) has been extensively applied in military, agriculture, geology, ecology, oceanography, etc., due to its prominent advantages of all-weather and all-time working capacity. Especially, in the marine field, SAR can provide numerous high-quality services for fishery management, traffic control, sea-ice monitoring, marine environmental protection, etc. Among them, ship detection in SAR images has attracted more and more attention on account of the urgent requirements of maritime rescue and military strategy formulation. Nowadays, most researches are focusing on improving the ship detection accuracy, while the detection speed is frequently neglected, regardless of traditional feature extraction methods or modern deep learning (DL) methods. However, the high-speed SAR ship detection is of great practical value, because it can provide real-time maritime disaster rescue and emergency military planning. Therefore, in order to address this problem, we proposed a novel high-speed SAR ship detection approach by mainly using depthwise separable convolution neural network (DS-CNN). In this approach, we integrated multi-scale detection mechanism, concatenation mechanism and anchor box mechanism to establish a brand-new light-weight network architecture for the high-speed SAR ship detection. We used DS-CNN, which consists of a depthwise convolution (D-Conv2D) and a pointwise convolution (P-Conv2D), to substitute for the conventional convolution neural network (C-CNN). In this way, the number of network parameters gets obviously decreased, and the ship detection speed gets dramatically improved. We experimented on an open SAR ship detection dataset (SSDD) to validate the correctness and feasibility of the proposed method. To verify the strong migration capacity of our method, we also carried out actual ship detection on a wide-region large-size Sentinel-1 SAR image. Ultimately, under the same hardware platform with NVIDIA RTX2080Ti GPU, the experimental results indicated that the ship detection speed of our proposed method is faster than other methods, meanwhile the detection accuracy is only lightly sacrificed compared with the state-of-art object detectors. Our method has great application value in real-time maritime disaster rescue and emergency military planning.

Real-Time Human Detection in Thermal Infrared Images

2021 ◽  
Author(s):  
Samah A. F. Manssor ◽  
Shaoyuan Sun ◽  
Mohammed Abdalmajed ◽  
Shima Ali
Keyword(s):  
Real Time ◽  
Network Architecture ◽  
Image Data ◽  
Human Detection ◽  
Accuracy Rate ◽  
Night Time ◽  
Method Performance ◽  
Thermal Infrared Images ◽  
Practical Model ◽  
Bounding Boxes

Abstract Human detection is a technology that detects pre-determined human shapes in the image and ignores everything else, which plays an irreplaceable role in video surveillance. However, modern person detectors have some inefficiencies in detecting pedestrians at night, and the accuracy rate is still insufficient. This paper presents a novel practical model for automatic real-time human detection at night-time. For this purpose, a new network architecture was proposed by improving the ting-yolov3 network for detecting pedestrians from TIR images based on the YOLO algorithm's tasks. The K-means clustering method clusters the image data, which contributes to obtaining excellent priority bounding-boxes. The proposed network was pre-trained on the original COCO dataset to obtain the initial weights. Through the comparison with the other three methods on the FLIR and DHU Night datasets showed that the proposed method performance was outperformed, in addition, to achieve a high score of accuracy (mAP%) in the TIR images. The method has a delay in detection time of 4.88ms. By improving the performance rates of human detection in TIR images, we expect this research to detect intruders in the night surveillance system.

Problems of ensuring metrological traceability of measurement results of indicators of quality for food products and food raw materials

2020 ◽  
pp. 64-70
Author(s):  
Mariya Y. Medvedevskikh ◽  
Anna S. Sergeeva
Keyword(s):  
Reference Materials ◽  
Mass Fraction ◽  
Nutritional Value ◽  
Animal Feed ◽  
Raw Materials ◽  
Meat Products ◽  
Metrological Traceability ◽  
Food Products ◽  
Freeze Dried ◽  
Measurement Results

The article raises the problem of ensuring metrological traceability of the measurement results of indicators of quality and nutritional value for food products and food raw materials: water (moisture), nitrogen (protein, crude protein), fat, ash and carbohydrates. The problem under consideration can be solved by applying reference materials of food composition, traceable to state primary measurement standards GET 173-2017 and GET 176-2019 and primary reference measurement procedures (PRMP), for attestation of measurement procedures and accuracy checking of measurement results. The article discusses the results of the PRMP development of mass fraction of fat, ash and carbohydrates in food products and food raw materials, as well as mass fraction of crude fat (oil content) in oil crops seeds and products based on them. The paper also presents metrological characteristics of reference materials of composition of dry dairy products, grain-milk dry porridges for nutrition of babies, grain dry porridges for nutrition of babies, egg powder, freeze-dried meat products, animal feed. The results of the work allow for building a chain of metrological traceability from GET 173-2017, GET 176-2019 and PRMP to routine measurement procedures, thereby ensuring the uniformity of measurements of nutritional value of food products.

High speed interferometric device for real-time correction of aero-optic effects

1995 ◽  
Author(s):  
Rod Clark ◽  
John Karpinsky ◽  
Gregg Borek ◽  
Eric Johnson
Keyword(s):  
Real Time ◽  
High Speed ◽  
Time Correction

Electrical Probing and Surface Imaging of Deep Sub-Micron Integrated Circuits

1999 ◽  
Author(s):  
Kenneth Krieg ◽  
Richard Qi ◽  
Douglas Thomson ◽  
Greg Bridges
Keyword(s):  
High Resolution ◽  
Integrated Circuits ◽  
Real Time ◽  
High Speed ◽  
Time Signal ◽  
Surface Imaging ◽  
Atomic Force ◽  
Submicron Structures ◽  
High Resolution Images ◽  
Capacitive Loading

Abstract A contact probing system for surface imaging and real-time signal measurement of deep sub-micron integrated circuits is discussed. The probe fits on a standard probe-station and utilizes a conductive atomic force microscope tip to rapidly measure the surface topography and acquire real-time highfrequency signals from features as small as 0.18 micron. The micromachined probe structure minimizes parasitic coupling and the probe achieves a bandwidth greater than 3 GHz, with a capacitive loading of less than 120 fF. High-resolution images of submicron structures and waveforms acquired from high-speed devices are presented.

Sign in / Sign up

Export Citation Format

Share Document