Computer vision based method for identification of freshness in mushrooms

Abstract

Scientific testing of mushroom (Agaricus Bisporus) samples, and especially non-destructive can prove to be very beneficial in different fields of research including agriculture, food processing, health care, etc. This is mainly because, when these mushrooms are left exposed in the open atmosphere, these tend to react with surrounding and develop a brown coloured pigment. This phenomenon is commonly referred to as the "enzymatic browning". Here, this reaction is utilized to observe certain chemical changes over a course of time. This has led to a result which can be used for classification of mushroom samples. Hence, this classification can be used to prove the gradual change in pattern behaviour of mushrooms during different time intervals. SVM classifier is seen to give the result with an accuracy of 80%. Thus, this classification can also give an overlook of whether these mushroom samples are fresh for consumption or not.