Design considerations for the processing system of a CNN-based automated surveillance system

Cameron, James A. D.; Savoie, Patrick; Kaye, Mary E.; Scheme, Erik

doi:10.1016/j.eswa.2019.06.037

Cited by 13 publications

(3 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…If convolution operations exhaust all the DSP resources, then the whole functioning of the applications will not work. In real-world applications, such as surveillance systems [23] and autonomous vehicles [4], an inference accelerator is only part of the applications, as the applications have other functional units. The proposed WRA-MF can balance the utilization of the on-chip resources by reducing the number of DSPs used in the convolution, both to increase the parallelism and efficiency of the convolutional computation itself and to use DSPs for more scenarios in which the use of multiplication cannot be avoided, such as average pooling, processing units other than neural network accelerators, and so on.…”

Section: Motivationmentioning

confidence: 99%

WRA-MF: A Bit-Level Convolutional-Weight-Decomposition Approach to Improve Parallel Computing Efficiency for Winograd-Based CNN Acceleration

Xiang,

Lv,

Meng

et al. 2023

Electronics

View full text Add to dashboard Cite

FPGA-based convolutional neural network (CNN) accelerators have been extensively studied recently. To exploit the parallelism of multiplier–accumulator computation in convolution, most FPGA-based CNN accelerators heavily depend on the number of on-chip DSP blocks in the FPGA. Consequently, the performance of the accelerators is restricted by the limitation of the DSPs, leading to an imbalance in the utilization of other FPGA resources. This work proposes a multiplication-free convolutional acceleration scheme (named WRA-MF) to relax the pressure on the required DSP resources. Firstly, the proposed WRA-MF employs the Winograd algorithm to reduce the computational density, and it then performs bit-level convolutional weight decomposition to eliminate the multiplication operations. Furthermore, by extracting common factors, the complexity of the addition operations is reduced. Experimental results on the Xilinx XCVU9P platform show that the WRA-MF can achieve 7559 GOP/s throughput at a 509 MHz clock frequency for VGG16. Compared with state-of-the-art works, the WRA-MF achieves up to a 3.47×–27.55× area efficiency improvement. The results indicate that the proposed architecture achieves a high area efficiency while ameliorating the imbalance in the resource utilization.

show abstract

Section: Motivationmentioning

confidence: 99%

WRA-MF: A Bit-Level Convolutional-Weight-Decomposition Approach to Improve Parallel Computing Efficiency for Winograd-Based CNN Acceleration

Xiang,

Lv,

Meng

et al. 2023

Electronics

View full text Add to dashboard Cite

show abstract

“…In recent years, the omnipresence of surveillance cameras in cities has motivated the development of automated monitoring systems [ 10 ]. Automated monitoring systems, using captured images, are expected to reduce the workload of operators and facilitate the analysis of a considerable amount of data [ 11 ]. The analytical results of significant amounts of image data are used in early warning systems for buildings [ 12 ], identifying the activity of workforces [ 13 ] and machines, and assessing productivity [ 14 ].…”

Section: Literature Reviewmentioning

confidence: 99%

Automated Sensing System for Real-Time Recognition of Trucks in River Dredging Areas Using Computer Vision and Convolutional Deep Learning

Chou

Liu

2021

Sensors

View full text Add to dashboard Cite

Sand theft or illegal mining in river dredging areas has been a problem in recent decades. For this reason, increasing the use of artificial intelligence in dredging areas, building automated monitoring systems, and reducing human involvement can effectively deter crime and lighten the workload of security guards. In this investigation, a smart dredging construction site system was developed using automated techniques that were arranged to be suitable to various areas. The aim in the initial period of the smart dredging construction was to automate the audit work at the control point, which manages trucks in river dredging areas. Images of dump trucks entering the control point were captured using monitoring equipment in the construction area. The obtained images and the deep learning technique, YOLOv3, were used to detect the positions of the vehicle license plates. Framed images of the vehicle license plates were captured and were used as input in an image classification model, C-CNN-L3, to identify the number of characters on the license plate. Based on the classification results, the images of the vehicle license plates were transmitted to a text recognition model, R-CNN-L3, that corresponded to the characters of the license plate. Finally, the models of each stage were integrated into a real-time truck license plate recognition (TLPR) system; the single character recognition rate was 97.59%, the overall recognition rate was 93.73%, and the speed was 0.3271 s/image. The TLPR system reduces the labor force and time spent to identify the license plates, effectively reducing the probability of crime and increasing the transparency, automation, and efficiency of the frontline personnel’s work. The TLPR is the first step toward an automated operation to manage trucks at the control point. The subsequent and ongoing development of system functions can advance dredging operations toward the goal of being a smart construction site. By intending to facilitate an intelligent and highly efficient management system of dredging-related departments by providing a vehicle LPR system, this paper forms a contribution to the current body of knowledge in the sense that it presents an objective approach for the TLPR system.

show abstract

“…Thus, eldercare should focus on this population. Using machine learning and contemporary computer vision techniques, we can detect things of interest more accurately than humans [5]. Hence, an automated eldercare system should be considered for good monitoring.…”

Section: Introductionmentioning

confidence: 99%

Mono Camera-based Human Skeletal Tracking for Squat Exercise Abnormality Detection using Double Exponential Smoothing

Hisham¹,

Hassan²,

Ibrahim³

et al. 2022

IJACSA

View full text Add to dashboard Cite

Human action analysis is an enthralling area of research in artificial intelligence, as it may be used to improve a range of applications, including sports coaching, rehabilitation, and monitoring. By forecasting the body's vital position of posture, human action analysis may be performed. Human body tracking and action recognition are the two primary components of video-based human action analysis. We present an efficient human tracking model for squat exercises using the open-source MediaPipe technology. The human posture detection model is used to detect and track the vital body joints within the human topology. A series of critical body joint motions are being observed and analysed for aberrant body movement patterns while conducting squat workouts. The model is validated using a squat dataset collected from ten healthy people of varying genders and physiques. The incoming data from the model is filtered using the double exponential smoothing method;the Mean Squared Error between the measured and smoothed angles is determined to classify the movement as normal or abnormal. Level smoothing and trend control have parameters of 0.8928 and 0.77256, respectively. Six out of ten subjects in the trial were precisely predicted by the model. The mean square error of the signals obtained under normal and abnormal squat settings is 56.3197 and 29.7857, respectively. Thus, by utilising a simple threshold method, the low-cost camera-based squat movement condition detection model was able to detect the abnormality of the workout movement.

show abstract

Design considerations for the processing system of a CNN-based automated surveillance system

Cited by 13 publications

References 15 publications

WRA-MF: A Bit-Level Convolutional-Weight-Decomposition Approach to Improve Parallel Computing Efficiency for Winograd-Based CNN Acceleration

WRA-MF: A Bit-Level Convolutional-Weight-Decomposition Approach to Improve Parallel Computing Efficiency for Winograd-Based CNN Acceleration

Automated Sensing System for Real-Time Recognition of Trucks in River Dredging Areas Using Computer Vision and Convolutional Deep Learning

Mono Camera-based Human Skeletal Tracking for Squat Exercise Abnormality Detection using Double Exponential Smoothing

Contact Info

Product

Resources

About