Research on Object Detection Algorithm Based on Neural Network Model Compression Technique
WEI Zhifei， SONG Quanhong， LI Fang， YANG Qingyu， WANG Aihua
Shanghai Electro-Mechanical Engineering Institute， Shanghai 201109， China
Abstract With the continuous development of deep learning in the field of object detection, the detection accuracy is constantly improved. However, the neural network algorithm model has high requirements on the computing resources of the hardware platform, so it is difficult to be applied in the embedded platform. In order to ensure that the neural network algorithm can meet the high accuracy and improve the efficiency of its operation, this paper carries out the research on the object detection algorithm based on neural network model compression technique. Firstly, K-means ++ clustering algorithm is used to cluster the prior box in the data set, in order to make algorithm have a good initial value in the initial stage of training. At the same time, aiming at the problem of computing speed, this paper optimizes the YOLOv3 which based on Darknet53 frame, and prunes the network model of YOLOv3.The experimental results show that the neural network model compression technique can improve the speed of the algorithm by two times with less loss of accuracy.
Key words ：
object detection
neural network
K-means++clustering algorithm
model pruning
Received: 08 October 2021
Published: 24 December 2021
[1]
GAO Nan, HU Ankang, HOU Lixun, CHANG Xin. An Rapid Prediction Method for Propeller Hydrodynamic Performance Based on Deep Learning
[2]
SUN Qianyang, ZHOU Li, DING Shifeng, LIU Renwei, DING Yi. An Artificial Neural Network-Based Method for Prediction of Ice Resistance of Polar Ships
[3]
ZHANG Xiaoyu, DU Xiangrun, ZHANG Jialiang, TAN Panlong, YANG Shibo. Research on Infrared Image Object Detection Method Based on Deformable DETR
[4]
CHEN Xiao1,2 (陈潇), ZHANG Rui1,2 (张瑞), TANG Xinyi1,2 (汤心溢), QIAN Juan3∗ (钱娟). Prediction of Pediatric Sepsis Using a Deep Encoding Network with Cross Features
[5]
TIAN Yuanyuan (田圆圆), JIN Yanrui (金衍瑞), LI Zhiyuan (李志远), LIU Jinlei (刘金磊), LIU Chengliang∗ (刘成良). Weighted Heterogeneous Graph-Based Incremental Automatic Disease Diagnosis Method
[6]
XIA Yu, WANG Lei. Risk Assessment Method for Marine Engineering Project Bidding Based on Back Propagation Neural Network
[7]
LIU Yu, LEI Xuemei. A Structured Pruning Method Integrating Characteristics of MobileNetV3
[8]
ZHAN Ke, ZHU Renchuan. A CNN-LSTM Ship Motion Extreme Value Prediction Model
[9]
XUE Yongboa (薛永波)，LIU Zhaob (刘钊)， LI Zeyanga (李泽阳)，ZHU Pinga* (朱平). CT Image Segmentation Method of Composite Material Based on Improved Watershed Algorithm and U-Net Neural Network Model
[10]
WU Xingl(武星), ZHANG Qingfeng(张庆丰), WANG Jianjia(王健嘉), YAO Junfeng(姚骏峰), Guo Yike.(郭毅可). Multiple Detection Model Fusion Framework for Printed Circuit Board Defect Detection
[11]
SHANG Fancheng, LI Chuanqing, ZHAN Ke, ZHU Renchuan. Application of Improved LSTM Neural Network in Time-Series Prediction of Extreme Short-Term Wave
[12]
LI Qing, HUANGFU Yubin, LI Jiangyun, YANG Zhifang, CHEN Peng, WANG Zihan. UConvTrans:A Dual-Flow Cardiac Image Segmentation Network by Global and Local Information Integration
[13]
CAO Xiangang1,2(曹现刚)，LEI Zhuol*(雷卓)，LI Yanchuan1(李彦川), ZHANG Mengyuan1(张梦园),DUAN Xinyu1(段欣宇). Prediction Method of Equipment Remaining Life Based on Self-Attention Long Short-Term Memory Neural Network
[14]
TANG Zhen, HAO Lihua, FENG Jing. Online Estimation of Power Shortage in Power Systems Driven by Local Frequency Measurement Data
[15]
WAN Anping, YANG Jie, MIAO Xu, CHEN Ting, ZUO Qiang, LI Ke. Boiler Load Forecasting of CHP Plant Based on Attention Mechanism and Deep Neural Network
2021, Vol. 4 
