Abstract Unmanned Aerial Vehicle (UAV) inspection has become one of main methods for current transmission line inspection, but there are still some shortcomings such as slow detection speed, low efficiency, and inability for low light environment. To address these issues, this paper proposes a deep learning detection model based on You Only Look Once (YOLO) v3. On the one hand, the neural network structure is simplified, that is the three feature maps of YOLO v3 are pruned into two to meet specific detection requirements. Meanwhile, the K-means++ clustering method is used to calculate the anchor value of the data set to improve the detection accuracy. On the other hand, 1000 sets of power tower and insulator data sets are collected, which are inverted and scaled to expand the data set, and are fully optimized by adding different illumination and viewing angles. The experimental results show that this model using improved YOLO v3 can effectively improve the detection accuracy by 6.0%, flops by 8.4%, and the detection speed by about 6.0%.

References

[1] Bhola R., Krishna N., Ramesh K., Senthilnath J., and Anand G., “Detection of the Power Lines in UAV Remote Sensed Images Using Spectral- Spatial Methods,” Journal of Environmental Management, vol. 206, pp. 1233-1242, 2018.

[2] Chen B. and Miao X., “Distribution Line Pole Detection and Counting Based on YOLO Using UAV Inspection Line Video,” Journal of Electrical Engineering and Technology, vol. 15, no. 1, pp. 441-448, 2020.

[3] Chen H., He Z., Shi B., and Zhong T., “Research on Recognition Method of Electrical Components Based on YOLO V3,” IEEE Access, vol. 7, pp. 157818-157829, 2019.

[4] Fang F., Li L., Zhu H., and Lim J., “Combining Faster R-CNN and Model-Driven Clustering for Elongated Object Detection,” IEEE Transactions on Image Processing, vol. 29, no. 1, pp. 2052- 2065, 2020.

[5] Hsu W. and Lin W., “Ratio-and-Scale-Aware YOLO for Pedestrian Detection,” IEEE Transactions on Image Processing, vol. 30, pp. 934-947, 2021.

[6] Ju M., Luo H., Wang Z., Hui B., and Chang Z., “The Application of Improved YOLO V3 in Multi-Scale Target Detection,” Applied Sciences, vol. 9, pp. 1-14, 2019.

[7] Kim K., Kim P., Chung Y., and Choi D., “Multi- Scale Detector for Accurate Vehicle Detection in Traffic Surveillance Data,” IEEE Access, vol. 7, pp. 78311-78319, 2019.

[8] Liu Z., Wang X., and Liu Y., “Application of Unmanned Aerial Vehicle Hangar in Transmission Tower Inspection Considering the Risk Probabilities of Steel Towers,” IEEE Access, vol. 7, pp. 159048-159057, 2019.

[9] Lu X., Ji J., Xing Z., and Miao Q., “Attention and Feature Fusion SSD for Remote Sensing Object Detection,” IEEE Transactions on Instrumentation and Measurement, vol. 70, pp. 1-9, 2021.

[10] Miao X., Liu X., Chen J., Zhuang S., Fan J., and Jiang H., “Insulator Detection in Aerial Images for Transmission Line Inspection Using Single Shot Multibox Detector,” IEEE Access, vol. 7, pp. 9945-9956, 2019.

[11] Maghawry A., Omar Y., and Badr A., “Self- Organizing Map vs Initial Centroid Selection Optimization to Enhance K-Means with Genetic Algorithm to Cluster Transcribed Broadcast News Documents,” The International Arab Journal of Information Technology, vol. 17, no. 3, pp. 316-324, 2020.

[12] Redmon J. and Farhadi A., “YOLOv3: An Incremental Improvement,” arXiv preprint arXiv:1804.02767, 2018.

[13] Ren S. and Zhang X., “Synthesizing Conjunctive and Disjunctive Linear Invariants by K-means++ and SVM,” The International Arab Journal of Information Technology, vol. 17, no. 6, pp. 847- 856, 2020.

[14] Tian Y., Yang G., Wang Z., Wang H., Li E., and Liang Z., “Apple Detection During Different Growth Stages in Orchards using the Improved 754 The International Arab Journal of Information Technology, Vol. 18, No. 6, November 2021 YOLO-V3 Model,” Computers and Electronics in Agriculture, vol. 157, pp. 417-426, 2019.

[15] Tian Y., Yang G., Wang Z., Li E., and Liang Z., “Detection of Apple Lesions in Orchards Based on Deep Learning Methods of CycleGAN and YOLOV3-Dense,” Journal of Sensors, vol. 2019, pp. 13, 2019.

[16] Wang S., Jiang F., Zhang B., Ma R., and Hao Q., “Development of UAV-Based Target Tracking and Recognition Systems,” IEEE Transactions on Intelligent Transportation Systems, vol. 21, no. 8, pp. 3409-3422, 2020.

[17] Wang K. and Liu M., “Object Recognition at Night Scene Based on DCGAN and Faster R- CNN,” IEEE Access, vol. 8, pp. 193168-193182, 2020.

[18] Xi D., Qin Y., Luo J., Pu H., and Wang Z., “Multipath Fusion Mask R-CNN with Double Attention and its Application into Gear Pitting Detection,” IEEE Transactions on Instrumentation and Measurement, vol. 70, pp. 1- 11, 2021.

[19] Zhang L., Liu J., Zhang B., Zhang D., and Zhu C., “Deep Cascade Model-Based Face Recognition: When Deep-Layered Learning Meets Small Data,” IEEE Transactions on Image Processing, vol. 29, pp. 1016-1029, 2020.

[20] Zhou L., Wei S., Cui Z., Fang J., Yang X., and Ding W., “Lira-YOLO: A Lightweight Model for Ship Detection in Radar Images,” Journal of Systems Engineering and Electronics, vol. 31, no. 5, pp. 950-956, 2020. Tu Renwei received his B.S. degrees from Jiaxing University, China, in 2017, and now he is a master’s student at Zhejiang Wanli University. His research interests mainly include digital video processing and application. Zhu Zhongjie received a Ph.D. degree in electronics science and technology from Zhejiang University, China, in 2004. He is currently a professor with the Faculty of Electronics and Information Engineering, Zhejiang Wanli University, China. His research interests mainly include video compression and communication, image analysis and understanding, watermarking and information hiding, and 3D image signal processing. Bai Yongqiang received his B.S. and M.S. degrees from Zhengzhou University, China, in 2006 and 2009 respectively, and received his Ph.D. degree from Ningbo University, China, in 2019. He is now a researcher in Zhejiang Wanli University, China. His research interests mainly include data hiding, digital watermarking and image processing. Gao Ming received M.S. degrees from Zhejiang University, China, in 2012. He is currently the executive director of Ninghai Power Supply Company Limited, State Grid Corporation of Zhejiang, China. His research interests mainly include power automation, artificial intelligence and its applications. Ge Zhifeng received B.S. degrees from Changsha University of Science and Technology, China, in 2005. He is currently the director of the development and construction department of Ninghai Power Supply Company Limited, State Grid Corporation of Zhejiang, China. His research interests mainly include power automation, artificial intelligence and its applications.