Xianqiao Chen

A novel approach for assistance with anti-collision decision making based on the International Regulations for Preventing Collisions at Sea

Marine intelligent anti-collision regulations have been a means of dealing with a particularly dangerous problem for many years. As the foundation for making anti-collision decisions, the International Regulations for Preventing Collisions at Sea formulated by the International Maritime Organization should always be considered. Based on the International Regulations for Preventing Collisions at Sea, the minimum distance required for anti-collision by only the give-way ship (under normal situations) and by both the give-way ship and the stand-on ship steering simultaneously (under critical situations) under all possible encounter situations are studied respectively. Rather than regarding ships as a point, a restricted area where no evasive action of other ships is allowed is introduced. Furthermore, the ship’s manoeuvrability is taken into account. The proposed model is also assessed by the traditional parameters used in anti-collision such as the closest point of approach, the distance to the closest point of approach and the time to the closest point of approach. The study shows that the results obtained in this paper are important and complement the above-mentioned regulations so that navigators can make wise decisions.

Research on image content retrieval with color features

The current retrieval system is mainly based on key words for image search, but in many case it cannot meet requirements for different user with different view. In this paper, a new image content retrieval system is presented, which is based on the image color features, and the effect that the color features have upon the result of image retrieval will be discussed. In this paper, the performance of new algorithm is also analyzed.

The Enhancement of Foggy Image with Useful Information Preservation

For ordinary degraded images, there have been many effective methods to correct them, one of which is the histogram equalization. But if histogram equalization or local histogram equalization approach is applied directly on foggy images in traffic, a special class of degraded images, it will lead some distortion. The degraded model based on scene depth estimation is an effective algorithm for this situation. However, the scene depth estimation is a difficult task and its computational complexity is high. In this paper, we present a new algorithm for foggy image restoration in traffic, which is based on histogram equalization. The useful information in low brightness area is preserved and the histogram of the sky area is translated and narrowed down. The enhancement area of useful information is enlarged and its computational complexity is low compared with the degraded model. KeywordsImage Enhancement; Image Restoration; Information Preservation; Foggy Image

Study on the Coordination between Berths and Channels in Tianjin Port

With the steady-state growth of throughput in Tianjin port, it is of great importance to search for the potential bottleneck that restricts its further development. The coordination development of berths and channel plays a vital role in the function exertion for every port. At first the statistical regularity of coming vessel, berthing time and so on are studied. Then the whole process for the vessels from entering the anchorage ground to leaving the port is realized by discrete-time simulation model, which is based on the proposed distributions. The capacity of the main channel in Tianjin port is studied under two scheduling strategies: First Come First Served (FCFS) and Minimum Berth-time First Served (MBFS). At last, some proposals are given to coordinate the development of berths and channel in Tianjin port. Simulation results show that MBFS is better than FCFS strategy and enhancing vessel grade is an effective way to promote the throughput of Tianjin port.

The Contrast Enhancement for Foggy Image Based on Atmospheric Degraded Model and Sky Separating

Foggy images in traffic is a special class of degraed images, and the enhancement or restoration of these image is very important for traffic safe. The histogram equalization or local histogram equalization is applied to these images directly will lead some distortion. The degraded model based on scene depth estimated is a effective alogrithm for this situation, but it depends on the scene depth estimation, which is a diffcult task. In this paper we present a new algorithm for foggy image restoration in traffic,which is improved based on histogram equalization, atmoshperic degraded model, sky area separating. Experiments show that the sky area separating is effective, enhancement is smoothing, and distortion is avoided with the new algorithm.