Search

Edge detection plays an immensely important role in image processing. In this paper, we propose a new model with the combination of the L 1 and L 2 fidelity terms on the basis of the well-known Mumford-Shah (MS) model. To solve this minimum model, we design an efficient algorithm based on a fixed-point iterative method and the Split-Bregman (SB) method. Experimental results show that the proposed model and algorithm can get better detected edges and have more advantages in efficiency and accuracy for different pure noisy images, even for mixed noisy images.

Sustainable management of end-of-life products is increasingly becoming a manufacturers' responsibility. Especially, under the duel drivers of legislative enforcement and economic profit, product recovery has been given greater attention by many manufacturers. However, the uncertainties within the reverse flow leads to a more complicated experience in operating product recovery networks. This paper aims to develop identification of the uncertainties within reverse logistics, and examine the effect of them on economic performance by developing a system dynamic model. The simulation result suggests that expanding the reverse logistics network, i.e. increasing collection points, will lead to profit up to a point, with the extent of this profit dependant on the number of collection points. However, this must be combined with legislative enforcement. The combined cooperation of customers, (re-) manufacturers, and government will shape the reverse logistics system, and gear it to success.

Ultra high-rise buildings always regarded as a landmark of a city. But it also brings some fire safety problems due to crowded people, long evacuation distance, fast fire spread, etc. So the fire consequences are very serious, and it is also difficult to search the personnel in the building under emergency, which make the fire-rescue action of the ultra high-rise building very difficult. As a result, a set of effective information collection and analysis method are needed, which includes the collection of trapped personnel location information, fire information, evacuation routine, similar fire case in history and so on. There has not been a mature ultra high-rise building fire rescue system worldwide so far. In the study, we focused on several key technologies related to this problem, including indoor personnel locating, multisensory fire information fusion, fire source identification and so on. Based on the above technologies, the framework and functional design of an ultra high-rise building fire command and rescue system has been developed.

In 20th century and in early 21st century, there were some massive earthquakes happened in China. Rapid and accurate estimation and assessment of casualties caused by earthquake is essential basis for government and administrators to initialize the appropriate level of emergent plan. In this paper, a casualty prediction system for earthquake based on geographic information system has been developed. The system has integrated the population distribution information, Geodata, building vulnerability information, and it can give a quick estimation of possible casualties caused by the earthquake through successful application of geographic information system. The system can give reasonable estimation of earthquake casualty. The visualized output data will be helpful for the disastrous consequences analysis and decision making.