Portrait Intelligent Analysis Application and Algorithm Optimization in Video Investigation

doi:10.3969/j.issn.1671-1122.2020.05.011

Abstract

Abstract:

Intelligent portrait analysis refers to the structural and visual analysis of portraits in video or video, and intelligent identification of the target person’s gender, age, hairstyle, etc. This technology has extremely high application value in video reconnaissance. The early algorithm of portrait recognition is to manually extract features and learn low-level visual features to classify and learn different attributes. This model based on traditional methods is often not satisfactory. In the field of computer vision, neural networks learned from massive image data have richer information and feature extraction than traditional methods. This paper attempts to train neural network models to detect and recognize pedestrians through deep learning technology, and intelligently treat pedestrians with different clothes. Recognition has better robustness, improves the accuracy of video portrait recognition, and expands the artificial intelligence technology for identity recognition.

Key words: video detection, portrait recognition, deep learning, convolutional neural network

CLC Number:

TP309

ZHANG Leihua, HUANG Jin, ZHANG Tao, WANG Shengyu. Portrait Intelligent Analysis Application and Algorithm Optimization in Video Investigation[J]. Netinfo Security, 2020, 20(5): 88-93.

Figures/Tables 5

References 10

[1]	WANG Fang, JIN Chengcheng, ZHAO Wenwu . Application of Dynamic Portrait Analysis System in Public Security Actual Combat[J]. Police Technology, 2018,34(S):95-97.
	王芳, 金成城, 赵文武 . 动态人像分析系统在公安实战中的运用[J]. 警察技术, 2018,34(S):95-97.
[2]	CHEN Chaowu . Application of Smart Video in Public Security Business and Introduction of SVAC Standards[J]. China Public Safety, 2009,8(3):15-36.
	陈朝武 . 智能视频在公安业务中能应用及SVAC标准介绍[J]. 中国公共安全, 2009,8(3):15-36.
[3]	SHEN Peng . Applied Research on Combining Practical Techniques and Techniques of Video Investigation with Practical Image Technology[J]. Police Technology, 2018,34(S):84-87.
	沈鹏 . 视频侦查实用技战法与实用图像技术结合的应用研究[J]. 警察技术, 2018,34(S):84-87.
[4]	JIANG Rongfeng, HAN Chunjun, HUANG Ling . The Construction and Application of Portrait Bayonet in the Actual Combat of Police Affairs[J]. Police Technology, 2018,34(S):73-76.
	姜荣峰, 韩春军, 黄凌 . 人像卡口在警务实战中的建设与应用[J]. 警察技术, 2018,34(S):73-76.
[5]	YE Xueqiang, CHU Leimin . Application of Deep Learning Algorithm in Public Security Dynamic Face Recognition technology[J]. Police Technology, 34(S):118-120.
	叶学强, 褚磊民 . 深度学习算法在公安动态人脸识别技术中的应用[J]. 警察技术, 2018,34(S):118-120.
[6]	SUDOWE P, SPITZER H, LEIBE B . Person Attribute Recognition with a Jointly-Trained Holistic CNN Model [C]//IEEE. 2015 IEEE International Conference on Computer Vision Workshop(ICCVW), December 7-13, 2015, Santiago, Chile. New Jersey: IEEE, 2015: 45-52.
[7]	LI Dangwei, CHEN Xiaotang, HUANG Kaiqi . Multi-attribute Learning for Pedestrian Attribute Recognition in Surveillance Scenarios [C]//IEEE. 3rd IAPR Asian Conference on Pattern Recognition, November 3-6, 2015, Kuala Lumpur, Malaysia. New Jersey: IEEE, 2015: 111-115.
[8]	YANG Luwei, ZHU Ligeng, WEI Yichen , et al. Attribute Recognition from Adaptive Parts [C]//IEEE. The 29th IEEE Conference on Computer Vision and Pattern Recognition, June 26- July 1, 2016, Las Vegas, USA. New Jersey: IEEE, 2016: 1-11.
[9]	LIN T Y, GOYAL P, GIRSHICK R , et al. Focal Loss for Dense Object Detection[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2020,42(2):318-327.
[10]	WANG Lei . Exploration and Construction of Portrait Recognition Technology in Smart Communities[J]. Police Technology, 2018,34(S):30-33.
	王雷 . 人像识别技术在智慧小区的探索与建设[J]. 警察技术, 2018,34(S):30-33.

重采样（全部采用 SoftmaxWithLoss）		朝向	背包	单双肩包	手提包	帽子	性别	口罩	年龄	上身颜色	上身风格	下身颜色	下身风格	手持物品	眼镜
	F1-score	0.7	0.49	0.497	0.496	0.539	0.8	0.64	0.33	0.19	0.4	0.1	0.33	0.70	0.59
	Acc（准确率）	0.88	0.88	0.877	0.98	0.96	0.81	0.98	0.567	0.51	0.95	0.69	0.98	0.735	0.699
SofmaxWithLoss/ Focal-Loss	F1-score	0.91	0.75	0.69	0.8	0.78	0.87	0.85	0.67	0.82	0.79	0.62	0.85	0.8	0.82
SofmaxWithLoss/ Focal-Loss	Acc （准确率）	0.96	0.89	0.88	0.93	0.95	0.92	0.99	0.63	0.85	0.96	0.87	0.96	0.92	0.95

[1]	WU Jing, LU Tianliang, DU Yanhui. Generation of Malicious Domain Training Data Based on Improved Char-RNN Model [J]. Netinfo Security, 2020, 20(9): 6-11.
[2]	WANG Wenhua, HAO Xin, LIU Yan, WANG Yang. The Safety Evaluation and Defense Reinforcement of the AI System [J]. Netinfo Security, 2020, 20(9): 87-91.
[3]	WANG Xiangyi, ZHANG Jian. Abnormal Behavior Detection of Virtualization Platform Based on Image and Machine Learning [J]. Netinfo Security, 2020, 20(9): 92-96.
[4]	LIU Jing, ZHANG Xueqian, LIU Quanming. Research on Captcha Recognition of Lightweight Convolutional Neural Network with Gabor [J]. Netinfo Security, 2020, 20(7): 77-84.
[5]	Rong WANG, Chunguang MA, Peng WU. An Intrusion Detection Method Based on Federated Learning and Convolutional Neural Network [J]. Netinfo Security, 2020, 20(4): 47-54.
[6]	Yongheng XIE, Yubo FENG, Qingfeng DONG, Mei WANG. Research on Data Ingestion Method Based on Deep Learning [J]. Netinfo Security, 2019, 19(9): 36-40.
[7]	Chunguang MA, Yaoyao GUO, Peng WU, Haibo LIU. Review of Image Enhancement Based on Generative Adversarial Networks [J]. Netinfo Security, 2019, 19(5): 10-12.
[8]	Yong FANG, Guangxiatian ZHU, Luping LIU, Peng JIA. Research on Browser Fuzz Sample Generation Technology Based on Deep Learning [J]. Netinfo Security, 2019, 19(3): 26-33.
[9]	ZHU Haiqi, JIANG Feng. Research and Analysis of Anomaly Detection Technology for Operation and Maintenance Data in the Era of Artificial Intelligence [J]. Netinfo Security, 2019, 19(11): 24-35.
[10]	Dagao DUAN, Yongheng XIE, Xinxin GAI, Zhanbin LIU. A Rumor Detection Model Based on Neural Network [J]. Netinfo Security, 2017, 17(9): 134-137.
[11]	Hao REN, Senlin LUO, Limin PAN, Junfeng GAO. Research on the Algorithm of Short Text Representation Based on Graph Structure [J]. Netinfo Security, 2017, 17(3): 46-52.
[12]	Wei SONG, Dongliang ZHANG, Zhenguo QI, Nan ZHENG. A Violent Video Detection Method Based on 3D Convolutional Networks [J]. Netinfo Security, 2017, 17(12): 54-60.
[13]	Wei SONG, Pei YANG, Jing YU, Wei JIANG. Terrorist Video Detection Using Visual Semantic Concepts [J]. Netinfo Security, 2016, 16(9): 12-17.
[14]	Qing LEI, Lihua JING, Deming ZHAO, Jilong ZHENG. Research on the Technology of Gun Detection System for Android APP Videos Based on Deep Learning [J]. Netinfo Security, 2016, 16(9): 149-153.
[15]	Qing HU, Benyong LIU. Speaker Recognition Algorithm Based on Convolutional Neural Networks [J]. Netinfo Security, 2016, 16(4): 55-60.