Katsuji Imai

Recognition of printed sequential plural patterns using the 3-layered BP model with feedback connections

The authors propose a method for recognizing printed sequential plural patterns using the three-layered backpropagation (BP) model with feedback connections. It enables segmentation and recognition concurrently independent of the order of the individual patterns. Segmentation and recognition are carried out concurrently by parallel distributed processing. Experiments involved recognizing original charge coupled device (CCD) camera images where stamped three-figured numerals were placed. These numerals were on very small electronic components. Experimental results show the recognition ability of the proposed method under the following three bad conditions in the original CCD camera images, i.e., when the numerals have lacked parts, blurred parts, or shaded parts, when two numerals next to each other are in contact, and when lumped noises are also included.<<ETX>>

Recognition of letters in lateral printed strings using a three-layered BP model with feedback connections

Although there is a simple and fast method to separate and classify letters in lateral printed strings by regarding a valley of the projection of the original image as a gap of letters, this method cannot be applied to cases where adjacent letters are connected or large clutter is involved [1]. This paper proposes a method which separates and recognizes each letter even in such cases as described in the foregoing, using a three-layered BP model with feedback connections. The recognition experiments using the proposed method were carried out with lateral strings of five printed letters (capital A to E) binarized through a CCD camera. The results show that the method is superior to conventional methods when applied to a letter string recognition system used in the process of factory automation.