Misreading prevention unit

Abstract

Abstract PURPOSE: To reduce the rate of misreading of OCR, by providing the circuit in which the video signal is binarized with floating slice in response to the contrast of background and the comparison circuit binarized with the fixed slice level for processing. CONSTITUTION: The video signal is converted into digital signal Y of 16-value at the A/D converter 1, and this is converted into the video signals Z 1 , Z 2 of binary respectively at the binary circuit 2 and comparator 3. The circuit performs binary with the floating slice level changed according to the contrast of the background of character and the comparator 3 performs it with the fixed slice level in the level requested for the read-in as character. The signals Z 1 , Z 2 are stored in the memory 4, the signal Z 1 is fed to the segmentation circuit 7, counter 5 and character identifying circuit 9, and the signal Z 2 is fed to the counter 6. The number of black video elements for the character is counted at the counters 5, 6, and the output is compared for the magnitude between the ratio of the count value of the counters 5, 6 and the set value. When the ratio of the count value is less than the set value, the character recognition of the signal Z 1 is inhibited at the circuit 9 to prevent misreading. COPYRIGHT: (C)1980,JPO&Japio

Inventors

• 👤 Norio Kanemitsu

Bibliographic Details