Automated object counting with multiple user-defined criteria

Automated object counting applications track, identify and count objects in a bounded image region while providing fast and objective results with minimum errors. A majority of the systems developed, however, utilize a fixed design which limits their adaptability to new types of objects. This paper describes algorithms that can be used for automated object counting systems based on user-selected attributes. The attributes include color, size, shape, orientation and texture, where for each selected attribute, the user can vary the similarity level. The system developed is able to locate, identify and count objects in the input image. The system is also capable of separating touching objects prior to counting. Object detection is100% accurate for input images containing similar objects and98.89% in the case of different objects. The system is able to correctly detect an object through its boundary provided that the contrast between the object and background is high. It can be seen through the performance test on real-world objects that the numerical techniques used for object comparison through the user-defined criteria agree with the manual method of classifying and counting