3-D object recognition using an ultrasonic sensor array and neural networks
Cho Hyun-Chul, Keeseong Lee · 2003
3-D object recognition independent of translation and rotation is presented using an ultrasonic sensor array, invariant moment vectors and neural networks. Using invariant moment vectors of the acquired 16/spl times/8 pixel data of square, rectangular, cylindric and regular triangular blocks, 3-D objects can be classified by self organizing feature map neural networks. Invariant moment vectors are constant independent of translation and rotation. The recognition rates for the training and testing data were 96.2% and 92.3%, respectively.