One class support vector machine used for blind pixel detection

Zhang Dongge; Fu Yutian

doi:10.3788/IRLA201847.0404001

One class support vector machine(OCSVM) was applied to classify the pixels of the infrared detectors, and it can detect the blind pixels by the random scenes. The blind pixel detection algorithms were reviewed in the beginning, and the imbalance distribution of the normal pixels and blind pixel was discussed in the following. The infrared image sequence was used to set up the OCSVM models and calculate the super sphere parameters, when the support vectors were represented by the Lagrangian coefficients. The OCSVM was an unsupervised method to cluster the pixels by the changing gray level and the random scenes. The super sphere model built by OCSVM would be refreshed by the updating image sequence, while the Lagrangian coefficients of the support vectors were recorded, so the blind pixels could be eventually classified by the statistic results of the preceding coefficients series. The mid-wave infrared 320256 image sequence was taken as an example to illustrate the proposed method, and it got the same results as the black body calibration. It could conclude that the OCSVM used for the online modeling of the blind pixel detection of the infrared detectors is adaptive and self-refreshing, and it could improve the efficiency of the infrared system test.

One class support vector machine used for blind pixel detection

1. Key Laboratory of Infrared System Detection and Imaging Technology,Chinese Academy of Sciences,Shanghai 200083,China;

2. Shanghai Institute of Technical Physics of the Chinese Academy of Sciences,Shanghai 200083,China

Received Date: 2017-11-10

Rev Recd Date: 2017-12-18

Publish Date: 2018-04-25

Key words:

Abstract: One class support vector machine(OCSVM) was applied to classify the pixels of the infrared detectors, and it can detect the blind pixels by the random scenes. The blind pixel detection algorithms were reviewed in the beginning, and the imbalance distribution of the normal pixels and blind pixel was discussed in the following. The infrared image sequence was used to set up the OCSVM models and calculate the super sphere parameters, when the support vectors were represented by the Lagrangian coefficients. The OCSVM was an unsupervised method to cluster the pixels by the changing gray level and the random scenes. The super sphere model built by OCSVM would be refreshed by the updating image sequence, while the Lagrangian coefficients of the support vectors were recorded, so the blind pixels could be eventually classified by the statistic results of the preceding coefficients series. The mid-wave infrared 320256 image sequence was taken as an example to illustrate the proposed method, and it got the same results as the black body calibration. It could conclude that the OCSVM used for the online modeling of the blind pixel detection of the infrared detectors is adaptive and self-refreshing, and it could improve the efficiency of the infrared system test.

HTML

Reference (16)

[1]	Bai Junqi, Jiang Yiliang, Zhao Chunguang, et al. Blind pixel detection algorithm for infrared focal plane array detector[J]. Infrared Technology, 2011, 33(4):233-240. (in Chinese)
[2]	Hao Lichao, Huang Aibo, Lai Canxiong, et al. Discussion of reliability analysis on IRFPAs by bad pixel[J]. Infrared and Laser Engineering, 2016, 45(5):0504004. (in Chinese)
[3]	Yao Qinfen, Gu Guohua. A new algorithm of blind pixel detection for IRFPA[J]. Infrared Technology, 2012, 34(8):441-443. (in Chinese)
[4]	Zhang Honghui, Luo Haibo, Yu Xinrong, et al. Blind pixel detection algorithm for IRFPA by applying pixel's characteristics histogram analysis[J]. Infrared and Laser Engineering, 2014, 43(6):1807-1811. (in Chinese)
[5]	Yan Fei, Hou Qingyu. Algorithm of blind pixel detection based on multi statistical characteristic abnormity[J]. Infrared and Laser Engineering, 2014, 43(2):454-457. (in Chinese)
[6]	Zhang Qiaozhou, Gu Guohua, Chen Qian, et al. Real time blind pixel detection and compensation technology based on two point parameters and self-adaptive window[J]. Infrared Technology, 2013, 35(3):139-145. (in Chinese)
[7]	Kan Bohan, Yin Jinjian, Li Lingjie, et al. IR blind pixels detection algorithm based on adjustable threshold window[J]. Laser Infrared, 2014, 44(8):949-952. (in Chinese)
[8]	Huang Xi, Zhang Jianqi, Liu Delian. Algorithm of blind pixels adaptive detection and compensation for infrared image[J]. Infrared and Laser Engineering, 2011, 40(2):370-376. (in Chinese)
[9]	Leng Hanbing, Gong Zhendong, Xie Qingsheng, et al. Adaptive blind pixel detection and compensation for IRFPA based on fuzzy median filter[J]. Infrared and Laser Engineering, 2014, 44(3):821-826. (in Chinese)
[10]	Li Yulu, Su Lan, Chen Daqian, et al. Adaptive blind pixel detection algorithms based on stepwise search strategy[J]. Infrared Technology, 2016, 38(6):457-460. (in Chinese)
[11]	GB/T 17444-2013 The technical norms for measurement and test of characteristic parameters of infrared focal plane arrays[S]. Beijing:China Standards Press, 2014. (in Chinese)
[12]	Pan Zhisong, Chen Bin, Miao Zhimin, et al. Overview of study on one class classifiers[J]. Acta Electronica Sinca, 2009, 37(11):2496-2503. (in Chinese)
[13]	Yao Fuguang, Zhong Xianxin, Tang Xiangyang, et al. Mechanism and implementation of one class support vector machine in fast foreign real time recognition[J]. Optics and Precision Engineering, 2009, 17(4):937-942. (in Chinese)
[14]	Lin Hao, Zhao Jiewen, Chen Quansheng, et al. Identification of egg freshness using near infrared spectroscopy and one class support vector machine algorithm[J]. Spectroscopy and Spectral Analysis, 2010, 30(4):929-932. (in Chinese)
[15]	Zheng Guansheng, Wang Jiandong, Gu Bin, et al. Analysis of one class incremental support vector machine[J]. Journal of Nanjing University of Aeronautics and Astronautics, 2015, 47(1):113-118. (in Chinese)
[16]	Wang Hongbo, Zhao Guangzhou, Qi Donglian, et al. Fast incremental learning method for one class support vector machine[J]. Journal of Zhejiang University (Engineering Science), 2012, 46(7):1327-1332. (in Chinese)

One class support vector machine used for blind pixel detection

doi: 10.3788/IRLA201847.0404001

Abstract

References

Proportional views

通讯作者: 陈斌, bchen63@163.com

Article Metrics

Related

Proportional views