II. PROBLEM DEFINTION
In agricultural industry quality assessment of product is of main concern. Nowadays, the quality of grain seed has been determined manually through a visual inspection by experienced personals. So it requires high degree of accuracy, high level of quality as well as correctness for a nondestructive quality evaluation method to satisfy customer need. Machine vision proved to be an effective tool that could be used to replace human inspectors for reliable and consistent judgment in estimating and comparing quality of seeds.
Basmati rice (Oryza sativa L) seed can be normal, long or small in size and defected seed can be chalky or broken as shown in Fig. 1. The blue highlighted boundaries are normal, red colored ones are long and green highlighted boundaries are small seed. Brown highlighted boundaries are chalky seed and yellow colored ones are broken seed. Normal seeds are most important while quantifying quality. These seeds are selected after the processing of seeds using image analysis. If not properly selected then degradation of quality of rice may occur. This paper proposes a new method for counting the number of Basmati rice (Oryza sativa L) seeds in terms of size as well as defected seed using machine vision processes with non-destructive technique to quantify the quality of Basmati rice(Oryza sativa L) seeds.