A new method utilizing Blind Source Separation based on Time-Frequency distributions (TFBSS) and wavelet packet denosing was proposed to extract the Fetal ElectroCardioGram (FECG) signal. The original eight ElectroCardioGram (ECG) signals obtained from the thoracic and abdominal area of the pregnant woman were firstly processed to eight components by utilizing rearrangement TFBSS. Then the Maternal ECG (MECG) signal and noise components in eight components were set by zero and the rest components were reconstructed by using the mixing matrix. The FECG with noise could be extracted by separating the reconstructed result by using rearrangement TFBSS. Finally, the baseline shift and noise in FECG were suppressed by wavelet packet denoising technique. The FECG could be extracted even under the condition of the fetal QRS wave being partly and entirely overlapped with the maternal QRS wave in the abdominal composite signal. The experimental results show that the clear FECG can be extracted by utilizing the proposed method.

To compensate the gray level error in optical 4f system, a method for gray level error compensation based on histogram matching and Radial Basis Function (RBF) neural network was proposed. The nonlinear transformation of histogram between input and output images in optical 4f System was fitted by RBF neural network, then the optimal estimation of curve for histogram matching between input and output images was obtained. The gray level error compensation image was obtained by utilizing histogram matching according to the optimal estimation of curve for histogram matching. The average Peak Signal-to-Noise Ratio (PSNR) gain achieved was 2.96dB and the visual effect of images processed was improved by utilizing the proposed method in actual optical 4f system. The experimental results show the gray level error in optical 4f system can be compensated effectively and the precision of optical information processing was improved by the proposed method.

News video is composed of a series of news items. It is very important for content-based analysis, indexing retrieval of news video to detect and segment the news items accurately. Analyzing the structure feature of news video, this paper detected and segmented the story items in the news video by using multi-feature such as silent points, shot boundary, anchorperson and theme caption. Also the study used different news videos in our experiment. The experimental results show that the proposed algorithm has high detection accuracy rate, and it can accomplish the task of segmenting news stories.