Photodetectors, as core devices for optical signal-to-electrical signal conversion, are widely used in machine vision, biomedicine, environmental monitoring, and other fields. Their spectral response characteristics directly determine their application range and detection accuracy. However, traditional band-by-band scanning measurement methods have limitations such as long cycle time, high cost, and inability to adapt to on-orbit operation, making it difficult to meet the needs of efficient evaluation. This study aims to combine the advantages of machine learning technology to establish a data-driven evaluation method for the spectral response characteristics of photodetectors, in order to solve the bottleneck of traditional methods and improve evaluation performance. A high-precision data acquisition system is constructed based on Time-Sensitive Network (TSN) synchronization technology to acquire spectral response data. Support Vector Machine (SVM), K-Nearest Neighbor (KNN), Random Forest (RF), and deep learning algorithms are used to construct evaluation models, with responsivity and specific detectivity as the core evaluation parameters. The results show that the established model achieves evaluation accuracies of 92.5% and 89.3% for detectors A and B, respectively, with corresponding F1 values of 91.8% and 88.7%. The overall model accuracy is further improved to 94.2%. This research provides a new technological path for evaluating the spectral response of photodetectors, offering theoretical support for the optimized design and process improvement of optoelectronic devices. It also lays the foundation for performance evaluation of novel photodetectors such as those made of two-dimensional materials and perovskites, promoting the intelligent development of photodetector technology.

Shaoyi Sun.Research on Evaluation of Spectral Response Characteristics of Photodetectors Based on Machine Learning. Academic Journal of Computing & Information Science (2025), Vol. 8, Issue 10: 124-128. https://doi.org/10.25236/AJCIS.2025.081015.

[1] Jiang Yan, Gao Feng, Li Lin. Photodetector with broadband-narrowband dual-function detection mode [J]. Journal of Luminescence, 2024.

[2] Tan Qichuan, Zeng Peng, Yang Zheqi. Design of computationally reconstructed spectral detector based on perovskite filter array [J]. Progress in Laser & Optoelectronics, 2024.

[3] Ge Xinle, Gao Lei, Sun Ling. Preliminary study on spectral response function inversion method based on linear programming [J]. Journal of North China University of Technology, 2024.

[4] Tian Ting, Zhang Qing, Xu Wen. Research progress on the application of spectral technology in crop nutrient monitoring [J]. Jiangsu Agricultural Sciences, 2024.

[5] Liu Sisi, Zhang Bo, Li Xingying, Xin Lei, Ying Guangguo, Chen Chang'er. Research progress on the application of machine learning in environmental analysis and detection [J]. Journal of Analysis and Testing, 2024.

[6] Zhang Haibo, Ji Ting, He Jiayu, Shi Linlin, Li Guohui, Cui Yanxia. Fast-response vertical structure two-dimensional perovskite photodetector [J]. Progress in Laser & Optoelectronics, 2024.

[7] He Yifei, Yang Deren, Pi Xiaodong. Research progress on infrared photodetectors based on silicon two-dimensional layered material heterojunction [J]. Materials Reports, 2024.

[8] Yang Rui, Yang Sishuo, Qian Lingxuan. Study on the influence of oxygen partial pressure on the performance of a-Ga₂O₃-based solar-blind ultraviolet photodetector [J]. Opto-Electronic Engineering, 2024.

[9] Yuan Liu, Zhang Weiqi, Wang Jiaqi, Zhang Hanwen, Wang Yanliang, Jiang Yadong, Tai Huiling. Development of short-wave infrared organic light detection materials [J]. Journal of University of Electronic Science and Technology of China, 2024.

[10] Chen Jianian, Shen Honglie, Li Yufang, Zhang Jingzhe, Li Hechao, Zhang Wenhao. Self-powered Si PEDOT:PSS heterojunction photodetector based on inverted pyramid structure [J]. Journal of Nanjing University of Aeronautics and Astronautics, 2024.

[11] Wang Xiangyong, Wang Qianqian, Dai Lili. Key technologies of LiDAR and their performance requirements in assisted driving [J]. Practical Automotive Technology, 2024.

[12] Wang Tao, Wan Yuemin. Application of stand-alone photoelectric intelligent fire detection technology in ancient buildings in Tibetan areas [J]. Science and Technology Innovation, 2024.