Detection of Overlapped Bullet Holes Based on Improved Otsu's Thresholding Method

被引：0
|
作者：
Song Y. [1 ,2 ]
Wang Y. [3 ]
Du B. [3 ]
Wang J. [1 ,2 ]
Dong X. [1 ]
机构：
[1] School of Electronic and Information Engineering, Suzhou University of Science and Technology, Suzhou
[2] Science and Technology on Near-Surface Detection Laboratory, Wuxi
[3] Baicheng Ordnance Test Center of China, Jilin
来源：
Binggong Xuebao/Acta Armamentarii | 2022年 / 43卷 / 04期
关键词：
Image processing; Improved otsu method; Minimum bounding rectangle; Overlapped bullet holes detection;
D O I：
10.12382/bgxb.2021.0201
中图分类号：
学科分类号：
摘要：
The traditional automatic target-scoring system has a low detection rate for the overlapped bullet holes on the target plate hit by successive projectiles. Generally, the automatic target-scoringing system based on image processing recognizes the overlapped target holes as one bullet hole, which leads to a high missed detection rate. To solve this problem, a new method based on Otsu is presented. The bullet holes recognition method based on improved Otsu method is employed to segment the overlapped bullet hole image and its background, and then the overlapped bullet hole region is obtained from the segmented image by the two-pass algorithm. The minimum bounding rectangle method is used to position the center of the two bullet holes in the overlapped bullet hole region. The results show that, compared with the traditional Hough transform, the proposed method can be used to effectively position the center of the two bullet holes and calculate the center coordinates. The measured error of bullet holes is not more than 3 pixel for the overlapped holes of which the degree of overlap is less than 88.73%. The improved method is capable of reducing the computational time and improving the positioning accuracy of bullet hole center with a low computational complexity, which plays a significant role in improving the detection accuracy of target hole. © 2022, Editorial Board of Acta Armamentarii. All right reserved.
引用
收藏
页码：924 / 930
页数：6
相关论文
共 15 条
  • [11] WANG H Y, CHENG Y., Detection algorithm of chest bitmap based on spatio-temporal context information, Journal of Electronics and Information Technology, 42, 8, pp. 1959-1967, (2020)
  • [12] MA Y D, ZHOU S Y., Salt and pepper noise filtering algorithm based on connectivity detection, Chinese Journal of Liquid Crystals and Displays, 35, 2, pp. 167-172, (2020)
  • [13] ZHANG F Q, WANG G F, ZENG Q N, Et al., New algorithm for minimum enclosing rectangle of the object in the image region based on center-of-gravity principle, Infrared and Laser Engineering, 42, 5, pp. 1382-1387, (2013)
  • [14] OTSU N., A threshold selection method from gray-level histograms, IEEE Transactions on Systems, Man, and Cybernetics, 9, 1, pp. 62-66, (1979)
  • [15] WU Y Q, WU J M, ZHAN B C., An Effective method of threshold selection for small object image, Acta Armamentarii, 32, 4, pp. 469-475, (2011)
12