In order to solve the problem of insufficient frequency resolution, occurring at the infrared detection data processing by the pulse phase thermography (PPT), a new frequency domain transform method was proposed. The fast Fourier transform (FFT) was utilized to choose the narrow frequency band included within the characteristic frequency, this band was processed by the chirp Z transform (CZT), the refined frequency structure was obtained. The clear amplitude and phase image sequences were reconstructed from the pixel-by-point analysis results. Without increasing the sampling time, a higher frequency resolution was achieve than the one of the PPT. Through the comparison of the detection signal from an aluminum material specimen, the results show that the algorithm can effectively refine the selected frequency band, get more precise characteristic frequency and effectively reduce the spectral leakage errors.
�±���ķ A V,л�� R W.��ɢʱ���źŴ���[M].����:������ͨ��ѧ������,2001 Oppenheim A V,Schafer R W.Discrete-time signal processing[M].Xi-an:Xi-an JiaoTong University Press,2001(in Chinese)
Wu D,Salerno A,Schonbach B,et al.Phabe sensitive modulation thermography and its applications for NDE[C]//Wurzbach R,Burleigh D.Proc SPIE Thermosense XIX.Bellingham:SPIE,1997:176-182
��ף��,����Ԫ,��.����Chirp-Z�任���LFMCW�״�Ĳ�ྫ��[J].�źŴ���,2002,18(2):110-112 Chen Zhuming,Ding Yiyuan,Xiang Jingcheng.Improving range precision of LFMCW radar by chirp-Z transform [J].Signal Processing,2002,18(2):110-112 (in Chinese)
������,������.����ģ̬Ӧ������Ƶ����Ϣ�����Ľṹ������ʶ���о�[J].Ӧ�û����빤�̿�ѧѧ��,2008,16(5):719-725 Guo Huiyong,Li Zhengliang.An improved information method based on modal strain energy and frequency for structural multiple damage identification [J].Journal of Basic Science and Engineering,2008,16(5):719-725(in Chinese)