The detection of defects in yarn-dyed fabric is one of the most difficult problems among the present fabric defect detection methods. The difficulty lies in how to properly separate patterns, textures, and defects in the yarn-dyed fabric. In this paper, a novel automatic detection algorithm is presented based on frequency domain filtering and similarity measurement. First, the separation of the pattern and yarn texture structure of the fabric is achieved by frequency domain filtering technology. Subsequently, segmentation of the periodic units of the pattern is achieved by using distance matching function to measure the fabric pattern. Finally, based on the similarity measurement technology, the pattern’s periodic unit is classified, and thus, automatic detection of the defects in the yarn-dyed fabric is accomplished.

No abstract

Vehicle surveillance in complex dark traffic scenes has been a key research topic, as the background is dramatically altered due to the reflections from headlights on normal, snowy and rainy roads. Under dark conditions a vehicles' headlights and rear lights are used for foreground extraction. The presented algorithm provides several steps including the detection, pairing and tracking of the headlights and rear lights. Firstly, the headlights are automatically extracted by a novel approach called azimuthally-blur which uses the exponentially attenuating nature of reflected light. This approach is robust on highly reflective scenes because it makes the headlights orthogonal to the reflections. The headlights are then paired by partitioning the image into subgroups such that in each group the headlights remain equidistant. The optimized tracker based on the Maximum a Posterior Probability (MAP) estimator is employed for further analysis such as speed estimation. This whole scheme is computationally inexpensive and can be deployed in ASICs. The proposed approach has outperformed state of the art methods in challenging unlit traffic scenes.Index Terms-Azimuthally-blur, false positives, maximum a posterior probability (MAP) estimator, night traffic videos, vehicle surveillance.

No abstract

No abstract