Due to the intensity distribution of X-ray image of thick steel pipe is not uniform, the contrast is low, the noise is big, and the size, shape, position and contrast of the blowholes defects are different, it is difficult to detect various types of blowholes automatically. Aiming at the problems that the traditional defect detection algorithm has a large workload of manually marking defect data, and the edge of the weld is difficult to accurately extract and other issues, a new unsupervised learning algorithm was proposed for the detection of various blowholes defects. Firstly, fast Independent Component Analysis (ICA) was used to learn a set of independent base vectors from the steel pipe X-ray image set, and a linear combination of the base vectors was used to selectively reconstruct the test image with blowholes defect. Then, the test image was subtracted from its reconstructed image to obtain the difference image, and the various blowholes were separated from the difference image by global threshold. There were 320 images in the training set and 60 images in the test set. The average sensitivity and accuracy of the proposed algorithm were 90.5% and 99.7%. The experimental results show that the algorithm can accurately detect all kinds of blowholes defects without manual marking the data or extracting the edge of the weld.

To improve the accuracy, reliability and computing efficiency of lane recognition and departure warning algorithm, a new lane detection and departure warning framework based on the hyperbolic model was proposed. Firstly, lane edge points were obtained from preprocessed image by searching feature points, and a least square fitting method was used to identify hyperbolic model of lane. The confidence factor of identified lane model was evaluated by a confidence function according to the characteristics of lane model and points adjacent to lane, and the reliability was compared with a threshold value to extract lane line. Then, in view of the continuous changing characteristics of lane between consecutive frames, particle filter algorithm was used to search lane edge points near the lane model obtained in previous frame, and an updated lane was identified and evaluated by confidence factor. Finally, based on a hyperbolic lane model established in aforementioned procedure, a spatial and temporal warning model of lane departure was proposed in image coordination system. The proposed algorithm was implemented on the PC platform and experiments on lane were done. The experimental results show that the method possesses good performances in recognition accuracy (92%) and average processing speed (40ms/frame), which can meet the application requirements.