Determination of the Number of Objects through Weight and Image Processing

Estimating the number of identical objects is a fundamental recognition task that requires exhaustive human effort. Automating this task in a warehouse would reduce the human load significantly. In this paper, we propose a method to automatically detect the number of objects, requiring only low-tech hardware and low-processing software, in a manner that provides optimal performance for the needed tasks. This method combines the physical weight of objects and simple image processing that do not require complicated operations, such as the detection of objects and their locations or the extraction of object figures. The experimental results demonstrate the increased effectiveness of the proposed method compared to methods based on weight alone or on image processing alone.