|
|
|
| Optimal recognition algorithm for solid nuclear track images based on morphology and machine learning |
| ZHANG Ziyang1, FAN Shengnan1, LI Mengxue1, ZHOU Wenshan2, DENG Jun1 |
1. National Institute for Radiological Protection, China Center for Disease Control and Prevention, Beijing 100088 China;
2. Hubei Provincial Center for Disease Control and Prevention, Wuhan 430079 China |
|
|
|
|
Abstract Objective To propose a computer recognition algorithm for solid nuclear track images based on the machine learning method, and to realize the automatic, fast and accurate recognition of nuclear tracks and improve the efficiency of solid track image analysis. Methods Firstly, 143 images containing tracks were scanned by morphological method to determine the location of suspected tracks, and 1250 material images were captured. 50% of the material were selected as the training set and 30% as the validation set for training the machine learning model. Another 20% of the material were selected as the test set for testing the model recognition result. The algorithm code was written and trained based on the MATLAB software. Results The established solid track recognition algorithm had a strong recognition capability, and the recognition accuracy of the test set could reach 84.8%. The machine learning model program constructed by the algorithm could evolve continuously with the input of training data, further improving the accuracy. Conclusion Based on image morphology and machine learning, the track recognition algorithm was investigated, by which the automatic recognition of solid tracks was better realized. In the future, we will increase the data input of the model, optimize the algorithm, and improve the recognition accuracy, in order to provide a more accurate and efficient method for automatic image track recognition.
|
|
Received: 26 December 2021
|
|
|
|
|
|
[1] 李志灵, 陈波, 卓维海, 等. 固体核径迹蚀刻仿真模型研究进展[J]. 中国辐射卫生,2019,28(4):473-476. DOI: 10.13491/j.issn.1004-714X.2019.04.034
Li ZL, Chen B, Zhuo WH, et al. Research progress on etching simulation of the nuclear track formed in solid state detectors[J]. Chin J Radiol Health, 2019, 28(4): 473-476. DOI: 10.13491/j.issn.1004-714X.2019.04.034
[2] 曹磊, 邓君, 吴鹏, 等. 固体径迹法测量氡及子体的判断阈和探测限[J]. 中国辐射卫生,2020,29(3):246-248. DOI: 10.13491/j.issn.1004-714X.2020.03.011
Cao L, Deng J, Wu P, et al. Determination threshold and detection limits of Radon and its daughter by solid track method[J]. Chin J Radiol Health, 2020, 29(3): 246-248. DOI: 10.13491/j.issn.1004-714X.2020.03.011
[3] 张庆贤, 葛良全, 肖才锦. 数学形态学在核径迹图像识别中的应用[J]. 核电子学与探测技术,2008,28(1):195-197. DOI: 10.3969/j.issn.0258-0934.2008.01.049
Zhang QX, Ge LQ, Xiao CJ. Application of mathematical morphology in discrimination nuclear track images[J]. Nucl Electron Detect Technol, 2008, 28(1): 195-197. DOI: 10.3969/j.issn.0258-0934.2008.01.049
[4] 张劲峰. 固体核径迹图像测量关键技术研究[D]. 成都: 西南交通大学, 2017.
Zhang JF. Research on the key technologies of solid state nuclear track image measurement[D]. Chengdu: Southwest Jiaotong University, 2017.
[5] 袁野. Matlab可视化与机器学习课程的案例教学实践[J]. 福建电脑,2019,35(7):116-118. DOI: 10.16707/j.cnki.fjpc.2019.07.040
Yuan Y. Case teaching practice of Matlab visualization and machine learning course[J]. J Fujian Comput, 2019, 35(7): 116-118. DOI: 10.16707/j.cnki.fjpc.2019.07.040
[6] 弟宇鸣, 叶红兵, 邱晓林, 等. 基于主成分变换的核径迹图像噪声分析及消除[J]. 核科学与工程,2007,27(1):37-40,19. DOI: 10.3321/j.issn:0258-0918.2007.01.008
Di YM, Ye HB, Qiu XL, et al. The analysis and removal of the nuclear track image noise based on principal components transform[J]. Chin J Nucl Sci Eng, 2007, 27(1): 37-40,19. DOI: 10.3321/j.issn:0258-0918.2007.01.008
[7] 范敦煌, 卓维海, 陈波. 固体核径迹自动识别系统技术概要[J]. 中国辐射卫生,2016,25(5):638-640. DOI: 10.13491/j.cnki.issn.1004-714X.2016.05.055
Fan DH, Zhuo WH, Chen B. Technical outline of solid nuclear track automatic recognition system[J]. Chin J Radiol Health, 2016, 25(5): 638-640. DOI: 10.13491/j.cnki.issn.1004-714X.2016.05.055
[8] 弟宇鸣, 叶红兵, 邱晓林, 等. 基于MATLAB核径迹图像聚焦算法的研究[J]. 核科学与工程,2006,26(4):316-320. DOI: 10.3321/j.issn:0258-0918.2006.04.006
Di YM, Ye HB, Qiu XL, et al. Study of an nuclear track image focalization arithmetic based on MATLAB[J]. Chin J Nucl Sci Eng, 2006, 26(4): 316-320. DOI: 10.3321/j.issn:0258-0918.2006.04.006
[9] 范胜男, 谭展, 王博, 等. 2013—2018年部分医院808台次医用磁共振成像设备影像质量检测与临床照片评估[J]. 中国辐射卫生,2020,29(6):632-636. DOI: 10.13491/j.issn.1004-714X.2020.06.014
Fan SN, Tan Z, Wang B, et al. Image quality test and clinical photographs evaluation of 808 medical magnetic resonance imaging systems in China during 2013—2018[J]. Chin J Radiol Health, 2020, 29(6): 632-636. DOI: 10.13491/j.issn.1004-714X.2020.06.014
[10] 宫法明, 刘芳华, 李厥瑾, 等. 基于深度学习的场景文本检测与识别[J]. 计算机系统应用,2021,30(8):179-185. DOI: 10.15888/j.cnki.csa.008038
Gong FM, Liu FH, Li JJ, et al. Scene text detection and recognition based on deep learning[J]. Comput Syst Appl, 2021, 30(8): 179-185. DOI: 10.15888/j.cnki.csa.008038
[11] Otsu N. A threshold selection method from gray-level histograms[J]. IEEE Trans Syst, Man, Cybern, 1979, 9(1): 62-66. DOI: 10.1109/TSMC.1979.4310076
[12] 吴冰, 秦志远. 自动确定图像二值化最佳阈值的新方法[J]. 测绘学院学报,2001,18(4):283-286. DOI: 10.3969/j.issn.1673-6338.2001.04.014
Wu B, Qin ZY. New approaches for the automatic selection of the optimal threshold in image binarization[J]. J Inst Surv Mapp, 2001, 18(4): 283-286. DOI: 10.3969/j.issn.1673-6338.2001.04.014
[13] 梁霄. 机器学习在量子物理学中的应用[D]. 合肥: 中国科学技术大学, 2019.
Liang X. Applications of machine learning in quantum physics[D]. Hefei: University of Science and Technology of China, 2019.
[14] 宋天阳. 基于遗传神经网络的全球年均气温预测研究[D]. 北京: 华北电力大学(北京), 2020. DOI: 10.27140/d.cnki.ghbbu.2020.000900.
Song TY. Research on global average annual temperature prediction based on genetic neural network[D]. Beijing: North China Electric Power University (Beijing), 2020. DOI: 10.27140/d.cnki.ghbbu.2020.000900.
[15] Mishkin D, Sergievskiy N, Matas J. Systematic evaluation of CNN advances on the ImageNet[EB/OL]. (2016-01-13)[2022-02-02]. https://arxiv.org/abs/1606.02228v1.
[16] 叶建龙, 胡新海. 基于卷积神经网络的图像识别算法研究[J]. 安阳师范学院学报,2021(5):14-18. DOI: 10.16140/j.cnki.1671-5330.2021.05.005
Ye JL, Hu XH. Research on image recognition algorithm based on convolution neural network[J]. J Anyang Norm Univ, 2021(5): 14-18. DOI: 10.16140/j.cnki.1671-5330.2021.05.005
|
|
|
|
