Abstract
We tried a method of grain sizing by image processing which is available to survey and analyze in short time. The high-efficiency method actualizes high spatial resolution information of grain size distribution. Thus, the information has a vailability to express a situation of stream flow better than traditional grain sizing methods. For this reason, we paid attention to 50 m reservoir area upper from the check dam in mountainous region and surveyed the grain distribution at 26 sites and river channel landform. The grain sizing by image processing provided the appropriate result qualitatively. Moreover we estimated the critical diameter of moving from hydraulic information simultaneously. A qualitative appropriate result is showed less than 50 mm error as a result, however, quantitative response is not found between the critical diameter of moving and the grain size distribution surveyed. Meanwhile,the different grain sizing methods that are image processing and traditional sieving are used to cover the bilateral weak point. Thereby, a peak of grain existence probability is found in the threshold diameter between image processing and sieving. This result indicates that it is necessary to change the threshold diameter much larger than the limit of image processing grain sizing.
