Abstract

The present work deals with the recognition of Indian postal letter sorting system based on PIN code. The optical camera catches the front and back view of each postal letter at a time. The vision system ensures from address and to address along with postal department stamp, seal etc., In this paper, an attempt has been made to recognize pin codes by using Connected Component (CC) approach, Artificial Neural Networks (ANN) and Barcode approaches. The six nearest neighbour CCs technique instead of 4 or 8 CCs has been adopted to recognize handwritten numerals. The ANN classifier technique has been used to recognize the numerals in the pin code. In this paper, the development of exact barcode to each pin code is introduced for sorting the postal letters automatically. The letters travel up to barcode scanner, if the PIN code is written on the envelope, equivalent barcodes are developed and printed on the each of the postal letters in the bottom line by the barcode printer, if PIN codes are not written on the envelope, by comparing a lookup table of place and PIN code, equivalent barcode to pin code will be printed on the bottom of the postal letter. Then the automatic letter sorting system is employed to sort the letters using the barcodes. The conveyor belt which is having gates at each check points works on the basis of on/off condition, enables the letter to travel up to exact location of the destination box. A comparative study has been made for the above approaches and the results are displayed. The experiments were performed on automatic postal letter sorting machine which is situated near to Meenambakkam Airport, Chennai. The experimental results reveal that barcode approach yields 99.5% of accuracy.

Key words: Pre-processing, classification, smoothing, neighboring pixels, barcode, pincode.