Wide area estimation of piled logs through image segmentation

Log-based industries especially paper manufacturers usually store large quantity of logs in wide opened area which is not good situation to automatically estimate the remained amount. Though the cross-section image can be utilized to effectively assess quantity of piled logs, investing cameras to view cross-section of every pile is quite inefficient. In this paper, the method for estimating quantity of piled logs is presented based on segmentation of image through grey scale morphological segmentation and histogram projection cut. The experiments are conducted based on two image captive devices which are surveillance and D-SLR cameras. The results confirm the usability of the proposed method with 78% accurate estimation.