Development of control system for fruit classification based on convolutional neural network

Recognition of visual images is one of the most important components of the control systems, information processing systems and automated decision-making systems. In this paper the approach to develop control system of objection recognition based on convolutional neural network (CNN) is considered. The CNN is applied to the tasks of fruits detection and recognition through parameter optimization. The result of our test had accuracy in the classification close to 94% for the 30 classes of 971 images, indicating that the proposed system and methods could be used for vision subsystem based control applications.