DETR-crowd is all you need
DOI:
https://doi.org/10.47813/2782-2818-2023-3-2-0213-0224Аннотация
"Crowded pedestrian detection" is a hot topic in the field of pedestrian detection. To address the issue of missed targets and small pedestrians in crowded scenes, an improved DETR object detection algorithm called DETR-crowd is proposed. The attention model DETR is used as the baseline model to complete object detection in the absence of partial features in crowded pedestrian scenes. The deformable attention encoder is introduced to effectively utilize multi-scale feature maps containing a large amount of small target information to improve the detection accuracy of small pedestrians. To enhance the efficiency of important feature extraction and refinement, the improved EfficientNet backbone network fused with a channel spatial attention module is used for feature extraction. To address the issue of low training efficiency of models that use attention detection modules, Smooth-L1 and GIOU are combined as the loss function during training, allowing the model to converge to higher precision. Experimental results on the Wider-Person crowded pedestrian detection dataset show that the proposed algorithm leads YOLO-X by 0.039 in AP50 accuracy and YOLO-V5 by 0.015 in AP50 accuracy. The proposed algorithm can be effectively applied to crowded pedestrian detection tasks.
Библиографические ссылки
Hu, W., Liu, X., & Xie, Z. (2022). ORE IMAGE SEGMENTATION APPLICATION BASED ON DEEP LEARNING AND GAME THEORY. In WORLD SCIENCE: PROBLEMS AND INNOVATIONS (pp. 71-76).
Zhouyi, X., Weijun, H., & Yanrong, H. (2022). Intelligent acquisition method of herbaceous flowers image based on theme crawler, deep learning and game theory. Кронос, 7(4 (66)), 44-52. DOI: https://doi.org/10.52013/2658-7556-66-4-12
Xie, Z., Hu, W., Fan, Y., & Wang, Y. (2022). RESEARCH ON MULTI-TARGET RECOGNITION OF FLOWERS IN LANDSCAPE GARDEN BASED ON GHOSTNET AND GAME THEORY. In РАЗВИТИЕ НАУКИ, ТЕХНОЛОГИЙ, ОБРАЗОВАНИЯ В XXI ВЕКЕ: АКТУАЛЬНЫЕ ВОПРОСЫ, ДОСТИЖЕНИЯ И ИННОВАЦИИ (pp. 46-56).
Song, Y., Chen, B., Liu, X., Weijun, H., Xiangyu, X., & Yuqi, Y. (2022). Audio and video editing system design based on OpenCV. Информатика. Экономика. Управление/Informatics. Economics. Management, 1(2), 0101-0120. DOI: https://doi.org/10.47813/2782-5280-2022-1-2-0101-0120
Xiaomin, L., Yuehang, S., Borun, C., Xiaobin, L., & Weijun, H. (2022). A novel deep learning based multi-feature fusion method for drowsy driving detection. Industry and agriculture, 34-49.
Hu, W., Zheng, T., Chen, B., Jin, J., & Song, Y. (2022). Research on product recommendation system based on deep learning. In Фундаментальные и прикладные научные исследования: актуальные вопросы, достижения и инновации (pp. 116-124).
He, W., Hu, W., Yang, Y., Shen, H., Wu, Y., Song, Y., & Liu, X. (2022). IMPROVED LEFT AND RIGHT HAND TRACKER USING COMPUTER VISION. In СТУДЕНЧЕСКИЕ НАУЧНЫЕ ИССЛЕДОВАНИЯ (pp. 21-29).
Xie, Z., Hu, W., Zhu, J., Li, B., Wu, Y., He, W., & Liu, X. (2022). LEFT AND RIGHT HAND TRACKER BASED ON CONVOLUTIONAL NEURAL NETWORK. In Актуальные вопросы современной науки образования (pp. 61-67).
He, W., Hu, W., Wu, Y., Sun, L., Liu, X., & Chen, B. (2022). DEVELOPMENT HISTORY AND RESEARCH STATUS OF CONVOLUTIONAL NEURAL NETWORKS. In СТУДЕНЧЕСКИЙ НАУЧНЫЙ ФОРУМ 2022 (pp. 28-36).
Yuan, C., Liu, X., & Zhang, Z. (2021, May). The Current Status and progress of Adversarial Examples Attacks. In 2021 International Conference on Communications, Information System and Computer Engineering (CISCE) (pp. 707-711). IEEE. DOI: https://doi.org/10.1109/CISCE52179.2021.9445917
Liu, X., Liu, W., Yi, S., & Li, J. (2020, October). Research on Software Development Automation Based on Microservice Architecture. In Proceedings of the 2020 International Conference on Aviation Safety and Information Technology (pp. 670-677). DOI: https://doi.org/10.1145/3434581.3434711
Liu, X., Xie, X., Hu, W., & Zhou, H. (2022). THE APPLICATION AND INFLUENCING FACTORS OF COMPUTER VISION: FOCUS ON HUMAN FACE RECOGNITION IN MEDICAL FIELD. In Наука, образование, инновации: актуальные вопросы и современные аспекты (pp. 32-37).
Shen, G., He, K., Jin, J., Chen, B., Hu, W., & Liu, X. (2022). CAPTURING AND ANALYZING FINANCIAL PUBLIC OPINION USING NLP AND DEEP FOREST. In НАУЧНЫЕ ИССЛЕДОВАНИЯ СТУДЕНТОВ И УЧАЩИХСЯ (pp. 66-71).
Chen, B., Song, Y., Cheng, L., He, W., Hu, W., Liu, X., & Chen, J. (2022). A REVIEW OF RESEARCH ON MACHINE LEARNING IN STOCK PRICE FORECASTING. In Наука и современное образование: актуальные вопросы, достижения и инновации (pp. 56-62).
He, K., Song, Y., Shen, G., He, W., & Liu, W. (2022). BASED ON DEEP REINFORCEMENT LEARNING AND COMBINED WITH TRENDS STOCK PRICE PREDICTION MODEL. In АКТУАЛЬНЫЕ ВОПРОСЫ СОВРЕМЕННЫХ НАУЧНЫХ ИССЛЕДОВАНИЙ (pp. 156-166).
Ou, S., Gao, Y., Zhang, Z., & Shi, C. (2021, December). Polyp-YOLOv5-Tiny: A Lightweight Model for Real-Time Polyp Detection. In 2021 IEEE 2nd International Conference on Information Technology, Big Data and Artificial Intelligence (ICIBA) (Vol. 2, pp. 1106-1111). IEEE. DOI: https://doi.org/10.1109/ICIBA52610.2021.9688145
Jiajun, J., & Wanting, Y. (2022). THE USE OF COMPUTER VISION TECHNOLOGY IN INTELLIGENT AGRICULTURAL MACHINERY. Наука и образование: сохраняя прошлое, создаём будущее, 9.
Xu, J., Chen, J., Li, B., & Li, X. (2022). ANALYSIS OF THE BARGAINING GAME AND BUYER'S BENEFIT MODEL. In СОВРЕМЕННЫЕ НАУЧНЫЕ ИССЛЕДОВАНИЯ: АКТУАЛЬНЫЕ ВОПРОСЫ, ДОСТИЖЕНИЯ И ИННОВАЦИИ (pp. 91-96).
Wu, J., Lee, P. P., Li, Q., Pan, L., & Zhang, J. (2018, May). CellPAD: Detecting performance anomalies in cellular networks via regression analysis. In 2018 IFIP Networking Conference (IFIP Networking) and Workshops (pp. 1-9). IEEE. DOI: https://doi.org/10.23919/IFIPNetworking.2018.8697027
Sun, Q., Zhao, C., Petrosian, O., & Li, Y. (2022). Power allocation in wireless cellular networks: stochastic algorithm based approach. Процессы управления и устойчивость, 9(1), 357-362.
Li , Q., Bai , M., Hu, W., He, J., He, K., Meng, L., & Xu, S. (2023). XXII International Scientific and Practical Conference. In Examination of the optimal solution problem for Go based on Monte Carlo algorithm (pp. 34–41). Melbourne; SCIENTIFIC VIEW. Retrieved March 26, 2023, from http://научныйвзгляд.рф.
Qingyuan, L., Wenke, D., Weijun, H., Kun, H., Weijia, L., Yanyou, W., Penghui, L.& Alina, R. (2023). OCTAVE programming for numerical analysis of free vibration of multi-degree-of-freedom structures. Industry and agriculture, 45-53.
Weijun, H., Weilong, H., Jipan, H., Kun, H.& Jialun, P. (2023).Game theoretic method and optimization of electric power companies. Industry and agriculture, 21-62.
Aircraft target detection vision system based on OpenCV / W. He, W. Liu, K. He, Ya. Wu // , 15 марта 2023 года, 2023. – P. 62-69. – EDN OAVTIS.
Jiajun, J., Yuehang, S., Geya, S., Borun, C., Kun, H., Weijia, L., ... & Weijun, H. (2022). The use of a discrete differential algorithm for deep learning has been the focus of research into the technologies around visual target tracking. Industry and agriculture, 6678.
Yu, X., Bo, L., & Xin, C. (2022). Low light combining multiscale deep learning networks and image enhancement algorithm. Modern Innovations, Systems and Technologies, 2(4), 0215–0232. https://doi.org/10.47813/2782-2818-2022-2-4-0215-0232 DOI: https://doi.org/10.47813/2782-2818-2022-2-4-0215-0232
Fan, K., Liu, W., He, K., Wang, Z., Ou, S., & Wu, Y. (2023). Review: the application of
artificial intelligence in distribution network engineering field. Информатика. Экономика.
Управление - Informatics. Economics. Management, 2(1), 0210–0218. https://doi.org/10.47813/2782-5280-2023-2-1-0210-0218
Ke, F., Chen-Yu, H., Weijia, L., Kun, H., Bin, S., & Yanyou, W. (2023). Research on computer vision application in industry field: focus on distribution network engineering. Современные инновации, системы и технологии - Modern Innovations, Systems and Technologies, 3(1), 0401–0409. https://doi.org/10.47813/2782-2818-2023-3-1-0401-0410 DOI: https://doi.org/10.47813/2782-2818-2023-3-1-0401-0410
Yuqi, Y., Wanting, Y., Xin, L., Jie, X., & Lihua, L. (2022). Studying electronic blood pressure monitor digital recognition algorithm based on computer vision and design. Modern Innovations, Systems and Technologies, 2(4), 0264–0277. https://doi.org/10.47813/2782-2818-2022-2-4-0264-0277 DOI: https://doi.org/10.47813/2782-2818-2022-2-4-0264-0277
He, K., Zhang, L., Liu, W., Fan, K., Song, P., & Wu, Y. (2023). The Review Of Application Of Deep Detection Network in Distribution Network Engineering field. In Современные стратегии и цифровые трансформации устойчивого развития общества, образования и науки (pp. 139-148).
Fan, K., Liu, W., He, K., Wang, Z., Ou, S., & Wu, Y. (2023). Review: the application of artificial intelligence in distribution network engineering field. Информатика. Экономика. Управление - Informatics. Economics. Management, 2(1), 0210–0218. https://doi.org/10.47813/2782- 5280-2023-2-1-0210-0218 DOI: https://doi.org/10.47813/2782-5280-2023-2-1-0210-0218
Загрузки
Опубликован
Как цитировать
Выпуск
Раздел
Лицензия
Copyright (c) 2023 Liu Weijia , Zishen Zheng , Ke Fan , Kun He , Taiqiu Huang , Weijia Liu , Xianlun Ke , Yuming Xu
Это произведение доступно по лицензии Creative Commons «Attribution» («Атрибуция») 4.0 Всемирная.
Журнал MIST - «Modern Innovations, Systems and Technologies» / «Современные инновации, системы и технологии» публикует материалы на условиях лицензии CreativeCommons Attribution 4.0 International (CC BY 4.0), размещенной на официальном сайте некоммерческой корпорации Creative Commons: 
This work is licensed under a Creative Commons Attribution 4.0 International License.
Это означает, что пользователи могут копировать и распространять материалы на любом носителе и в любом формате, адаптировать и преобразовывать тексты, использовать контент для любых целей, в том числе коммерческих. При этом должны соблюдаться условия использования — указание автора оригинального произведения и источника: следует указывать выходные данные статей, предоставлять ссылку на источник, а также указывать, какие изменения были внесены.
