基于邻域粗糙集和K近邻分类的核素识别方法研究

陈宸; 吴桓; 王国帆; 姚正勇

doi:10.3724/j.0253-3219.2026.hjs.49.250382

您当前的位置：

首页 >

文章列表页 >

基于邻域粗糙集和K近邻分类的核素识别方法研究

核电子学与仪器 | 更新时间：2026-02-13

- 基于邻域粗糙集和K近邻分类的核素识别方法研究
- Nuclide identification method based on neighborhood rough set and KNN classification
- 核技术 2026年49卷第2期文章编号：020401
- 作者机构：
  
  重庆建安仪器有限责任公司重庆 400060
- 作者简介：
  
  陈宸，男，1987年出生，2013年于重庆邮电大学获硕士学位，正高级工程师，从事人工智能及γ能谱分析
  吴桓，E-mail：wuh_1982@163.com
- 基金信息：
- DOI：10.3724/j.0253-3219.2026.hjs.49.250382
  中图分类号： TL81
- 收稿：2025-10-10，
  
  修回：2025-11-10，
  
  纸质出版：2026-02-15
- 稿件说明：
移动端阅览
陈宸,吴桓,王国帆等.基于邻域粗糙集和K近邻分类的核素识别方法研究[J].核技术,2026,49(02):020401.

CHEN Chen,WU Huan,WANG Guofan,et al.Nuclide identification method based on neighborhood rough set and KNN classification[J]. NUCLEAR TECHNIQUES,2026,49(02):020401.
陈宸,吴桓,王国帆等.基于邻域粗糙集和K近邻分类的核素识别方法研究[J].核技术,2026,49(02):020401. DOI： 10.3724/j.0253-3219.2026.hjs.49.250382. CSTR： 32193.14.hjs.CN31-1342/TL.2026.49.250382.

CHEN Chen,WU Huan,WANG Guofan,et al.Nuclide identification method based on neighborhood rough set and KNN classification[J]. NUCLEAR TECHNIQUES,2026,49(02):020401. DOI： 10.3724/j.0253-3219.2026.hjs.49.250382. CSTR： 32193.14.hjs.CN31-1342/TL.2026.49.250382.

摘要

针对γ能谱数据的高维特性和噪声干扰导致传统核素识别方法普遍存在识别精度不足与计算效率较低的问题，提出一种融合邻域粗糙集与K近邻分类的核素识别方法，以满足便携式核素识别设备在资源受限环境下的实际应用需求。该方法首先采用主成分分析法(Principal Component Analysis，PCA)对原始1 024维γ能谱数据进行降维，提取出累积贡献率≥99%的主成分，以消除冗余信息并抑制噪声干扰；随后引入邻域粗糙集的属性约简算法，通过定义样本邻域关系并计算属性依赖度，结合启发式前向贪心搜索策略对特征属性进行约简，最大限度保留关键判别信息并提升特征子集的分类鉴别能力；最后将约简后的低维特征输入K近邻分类器，依据距离加权投票机制计算各类核素的置信度，实现高效率、准确的核素识别。基于一套LaBr₃（Ce）探测系统与1 024道多道分析器对识别算法进行实验测试，采集包括12种单一核素及2种混合核素在内的224组γ能谱样本，在STM32F407ZGT6单片机硬件平台上进行测试，在邻域半径

=0.2时对包含特征峰重叠的混合核素在内的测试集的平均识别精度达到98.5%，单次核素识别运行时间控制在140 ms内，显著提升了核素识别的准确性与计算效率，为后续便携式核素识别仪的算法部署提供了可靠保障。

Abstract

Background

The high-dimensionality and inherent noise in gamma spectrum data often lead to low identification accuracy and poor computational efficiency in conventional nuclide identification methods. Therefore

a new method that can provide high accuracy and computational efficiency in portable devices is needed.

Purpose

This study aims to develop an efficient method for nuclide identification by combining neighborhood rough set (NRS) theory and K-nearest neighbor (KNN) classifier.

Methods

This method involved three-stage process. Firstly

principal component analysis (PCA) was applied to gamma spectrum data for dimensionality reduction and reduction of linear redundancy. Subsequently

a heuristic search strategy based on NRS was employed to optimize the feature subset by eliminating redundant attributes. Lastly

the KNN classifier was used for efficient nuclide identification in the reduced low dimensional feature space. In the meantime

the identification method was implemented on an STM32F407ZGT6 platform

and experimental verification was performed with a LaBr₃(Ce) detection system and a 1 024-channel multichannel analyzer.

Results

The experiment used a dataset of 224 gamma spectra

comprising 12 single-nuclide and 2 mixed-nuclide categories was collected in the experiment

and verification results show that an average identification accuracy of 98.5% was achieved by proposed method on the test set with a neighborhood radius of

=0.2. The processing time for a single identification is within 140 ms.

Conclusions

Experimental results confirm the superior generalization performance and high computational efficiency of the proposed method across various scenarios

including mixed-nuclide conditions

hence providing a reliable and efficient technical pathway for efficient nuclide identification in portable radiation detection devices.

关键词

Keywords

references

陈亮 , 魏义祥 , 屈建石 . 便携式γ谱仪中的核素识别算法 [J ] . 清华大学学报(自然科学版) , 2009 , 49 ( 5 ): 635 – 638 .

CHEN Liang , WEI Yixiang , QU Jianshi . Nuclide identification algorithm for a portable gamma spectrometer [J ] . Journal of Tsinghua University (Science and Technology) , 2009 , 49 ( 5 ): 635 – 638 .

Huang Y , Liu M Z , Liu X H , et al . Nuclide spectrum peak searching algorithm based on multiple morphological structuring elements [J ] . Journal of Physics: Conference Series , 2020 , 1634 ( 1 ): 012078 . DOI: 10.1088/1742-6596/1634/1/012078 http://dx.doi.org/10.1088/1742-6596/1634/1/012078 .

Luo R , Huang Y , Liu X H . Comparative study on peak-seeking methods of the mixed radioactive energy spectrum [J ] . OALib , 2020 , 7 ( 8 ): 1 – 10 . DOI: 10.4236/oalib. 1106676 http://dx.doi.org/10.4236/oalib.1106676 .

Jhung S , Hur S , Cho G , et al . A neural network approach for identification of gamma-ray spectrum obtained from silicon photomultipliers [J ] . Nuclear Instruments and Methods in Physics Research Section A: Accelerators, Spectrometers, Detectors and Associated Equipment , 2020 , 954 : 161704 . DOI: 10.1016/j.nima.2018.12.019 http://dx.doi.org/10.1016/j.nima.2018.12.019 .

Koo B T , Lee H C , Bae K , et al . Development of a radionuclide identification algorithm based on a convolutional neural network for radiation portal monitoring system [J ] . Radiation Physics and Chemistry , 2021 , 180 : 109300 . DOI: 10.1016/j.radphyschem.2020. 109300 http://dx.doi.org/10.1016/j.radphyschem.2020.109300 .

Zhao S Q , Zhou S M , Chen R . Research on radionuclide identification method based on GASF and deep residual network [J ] . Applied Sciences , 2025 , 15 ( 3 ): 1218 . DOI: 10.3390/app15031218 http://dx.doi.org/10.3390/app15031218 .

杨馨 , 张坤明 , 陈涛 . 基于机器学习预测膨润土中阳离子的分配系数 [J ] . 核技术 , 2025 , 48 ( 6 ): 060302 . DOI: 10.11889/j.0253-3219.2025.hjs.48.240431 http://dx.doi.org/10.11889/j.0253-3219.2025.hjs.48.240431 .

YANG Xin , ZHANG Kunming , CHEN Tao . Machine learning-based prediction of cation distribution coefficients in bentonite [J ] . Nuclear Techniques , 2025 , 48 ( 6 ): 060302 . DOI: 10.11889/j.0253-3219.2025.hjs.48.240431 http://dx.doi.org/10.11889/j.0253-3219.2025.hjs.48.240431 .

Li F , Luo C Y , Wen Y Z , et al . A nuclide identification method of γ spectrum and model building based on the transformer [J ] . Nuclear Science and Techniques , 2025 , 36 ( 1 ): 7 . DOI: 10.1007/s41365-024-01564-5 http://dx.doi.org/10.1007/s41365-024-01564-5 .

Huang X , Yuan Y G , Zhu Y X , et al . A low-consumption multiple nuclides identification algorithm for portable gamma spectromete [J ] . Nuclear Science and Techniques , 2025 , 36 ( 7 ): 121 . DOI: 10.1007/s41365-025-01701-8 http://dx.doi.org/10.1007/s41365-025-01701-8 .

Candy J V , Breitfeller E , Guidry B L , et al . Physics-based detection of radioactive contraband: a sequential Bayesian approach [J ] . IEEE Transactions on Nuclear Science , 2009 , 56 ( 6 ): 3694 – 3711 . DOI: 10.1109/TNS.2009. 2034374 http://dx.doi.org/10.1109/TNS.2009.2034374 .

万琳健 , 张璇 , 张春雷 , 等 . 基于序贯贝叶斯方法的核素快速检测方法研究 [J ] . 核技术 , 2025 , 48 ( 1 ): 010404 . DOI: 10.11889/j.0253-3219.2025.hjs.48.240190. CSTR: 32193.14.hjs.CN31-1342/TL.2025.48.240190 http://dx.doi.org/10.11889/j.0253-3219.2025.hjs.48.240190.CSTR:32193.14.hjs.CN31-1342/TL.2025.48.240190 .

WAN Linjian , ZHANG Xuan , ZHANG Chunlei , et al . Fast detection method for radionuclides based on Sequential Bayesian approach [J ] . Nuclear Techniques , 2025 , 48 ( 1 ): 010404 . DOI: 10.11889/j.0253-3219.2025.hjs.48.240190. CSTR: 32193.14.hjs.CN31-1342/TL.2025.48.240190 http://dx.doi.org/10.11889/j.0253-3219.2025.hjs.48.240190.CSTR:32193.14.hjs.CN31-1342/TL.2025.48.240190 .

曹仁龙 , 李孝哲 , 黄亭璋 , 等 . 一种用于核素识别的序贯贝叶斯方法 [J ] . 核电子学与探测技术 , 2025 , 45 ( 3 ): 343 – 352 . DOI: 10.20173/j.cnki.ned.20250110.001 http://dx.doi.org/10.20173/j.cnki.ned.20250110.001 .

CAO Renlong , LI Xiaozhe , HUANG Tingzhang , et al . A sequential Bayesian method for nuclide recognition [J ] . Nuclear Electronics & Detection Technology , 2025 , 45 ( 3 ): 343 – 352 . DOI: 10.20173/j.cnki.ned.20250110.001 http://dx.doi.org/10.20173/j.cnki.ned.20250110.001 .

罗庚 , 石睿 , 曾树鑫 , 等 . 基于核脉冲峰值序列二维卷积神经网络的核素识别方法 [J ] . 核技术 , 2025 , 48 ( 3 ): 030502 . DOI: 10.11889/j.0253-3219.2025.hjs.48.240301. CSTR: 32193.14.hjs.CN31-1342/TL.2025.48.240301 http://dx.doi.org/10.11889/j.0253-3219.2025.hjs.48.240301.CSTR:32193.14.hjs.CN31-1342/TL.2025.48.240301 .

LUO Geng , SHI Rui , ZENG Shuxin , et al . Radionuclide identification method of two-dimensional convolutional neural network based on nuclear pulse peak sequence [J ] . Nuclear Techniques , 2025 , 48 ( 3 ): 030502 . DOI: 10.11889/j.0253-3219.2025.hjs.48.240301. CSTR: 32193.14.hjs.CN31-1342/TL.2025.48.240301 http://dx.doi.org/10.11889/j.0253-3219.2025.hjs.48.240301.CSTR:32193.14.hjs.CN31-1342/TL.2025.48.240301 .

Li X Z , Zhang Q X , Tan H Y , et al . Fast nuclide identification based on a sequential Bayesian method [J ] . Nuclear Science and Techniques , 2021 , 32 ( 12 ): 143 . DOI: 10.1007/s41365-021-00982-z http://dx.doi.org/10.1007/s41365-021-00982-z .

Pawlak Z . Rough sets [J ] . International Journal of Computer & Information Sciences , 1982 , 11 ( 5 ): 341 – 356 . DOI: 10.1007/BF01001956 http://dx.doi.org/10.1007/BF01001956 .

陈宸 , 赵军 . 一种新的基于二进制分辨矩阵的属性约简方法 [J ] . 计算机应用与软件 , 2013 , 30 ( 9 ): 123 – 127 . DOI: 10.3969/j.issn.1000-386x.2013.09.035 http://dx.doi.org/10.3969/j.issn.1000-386x.2013.09.035 .

CHEN Chen , ZHAO Jun . A new attribute reduction method based on binary discernibility matrix [J ] . Computer Applications and Software , 2013 , 30 ( 9 ): 123 – 127 . DOI: 10.3969/j.issn.1000-386x.2013.09.035 http://dx.doi.org/10.3969/j.issn.1000-386x.2013.09.035 .

Yu B , Hu Y , Kang Y , et al . A novel variable precision rough set attribute reduction algorithm based on local attribute significance [J ] . International Journal of Approximate Reasoning , 2023 , 157 : 88 – 104 . DOI: 10.1016/j.ijar.2023.03.002 http://dx.doi.org/10.1016/j.ijar.2023.03.002 .

胡清华 , 于达仁 , 谢宗霞 . 基于邻域粒化和粗糙逼近的数值属性约简 [J ] . 软件学报 , 2008 , 19 ( 3 ): 640 – 649 . DOI: 10.3724/SP.J.1001.2008.00640 http://dx.doi.org/10.3724/SP.J.1001.2008.00640 .

HU Qinghua , YU Daren , XIE Zongxia . Numerical attribute reduction based on neighborhood granulation and rough approximation [J ] . Journal of Software , 2008 , 19 ( 3 ): 640 – 649 . DOI: 10.3724/SP.J.1001.2008.00640 http://dx.doi.org/10.3724/SP.J.1001.2008.00640 .

郑文彬 , 李进金 , 何秋红 . 基于属性重要度的变精度邻域粗糙集属性约简算法 [J ] . 计算机科学 , 2019 , 46 ( 12 ): 261 – 265 . DOI: 10.11896/jsjkx.181102184 http://dx.doi.org/10.11896/jsjkx.181102184 .

ZHENG Wenbin , LI Jinjin , HE Qiuhong . Attribute reduction algorithm for neighborhood rough sets with variable precision based on attribute importance [J ] . Computer Science , 2019 , 46 ( 12 ): 261 – 265 . DOI: 10.11896/jsjkx.181102184 http://dx.doi.org/10.11896/jsjkx.181102184 .

Wu S Z , Wang L T , Ge S Y , et al . Neighborhood rough set with neighborhood equivalence relation for feature selection [J ] . Knowledge and Information Systems , 2024 , 66 ( 3 ): 1833 – 1859 . DOI: 10.1007/s10115-023-01999-z http://dx.doi.org/10.1007/s10115-023-01999-z .

Qian D M , Liu K Y , Wang J , et al . Attribute reduction based on directional semi-neighborhood rough set [J ] . International Journal of Machine Learning and Cybernetics , 2025 , 16 ( 4 ): 2523 – 2535 . DOI: 10.1007/s13042-024-02406-x http://dx.doi.org/10.1007/s13042-024-02406-x .

Sun L , Zhang X Y , Xu J C , et al . An attribute reduction method using neighborhood entropy measures in neighborhood rough sets [J ] . Entropy , 2019 , 21 ( 2 ): 155 . DOI: 10.3390/e21020155 http://dx.doi.org/10.3390/e21020155 .

Biau G , Devroye L . Lectures on the Nearest Neighbor Method [M ] . Cham : Springer International Publishing , 2015 . DOI: 10.1007/978-3-319-25388-6 http://dx.doi.org/10.1007/978-3-319-25388-6 .

Li W , Zhang Y , Sun Y F , et al . Approximate nearest neighbor search on high dimensional data: experiments, analyses, and improvement [J ] . IEEE Transactions on Knowledge and Data Engineering , 2020 , 32 ( 8 ): 1475 – 1488 . DOI: 10.1109/TKDE.2019.2909204 http://dx.doi.org/10.1109/TKDE.2019.2909204 .

浏览量

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

基于深度神经网络的放射性废物桶γ能谱解析方法

便携式LaBr₃(Ce) γ谱仪稳谱技术研究

r放射性门式检测系统的蒙特卡罗模拟及设计

中子活化测量谱的计算机自动识别