FRUIT CATEGORIZATION AND DISEASE DETECTION USING ML

Abstract

In this paper we discuss about the fruit categorization and quality maintenance using raspberry pi board by leveraging the machine learning concepts. India is predominantly agro based economy with Agriculture as the main source of income for the farmers who are termed as the backbone of the country. Our farmers are known to use typical old methods and due to lack of education, they are still far from incorporating the advanced technical tools in agriculture. We are proposing a low cost yet powerful fruit quality maintenance device which can be helpful for our fruit merchants and farmers. Fruit Detection classification and categorization has been implemented in this paper using Machine learning and embedded concepts. We have selected Apple, Banana, Orange, Papaya etc. fruits for the demonstration. We studied the fruit detection by methods such Haar cascade classifier and tensor flow classifier. We trained the fruit classifier by using machine learning concepts and obtained the trained classifier to detect & categorize the fruits with quality. The electronic components used here are Raspberry Pi. Instead of raspberry pi, the laptop with Linux operating system (Ubuntu) can be used. Through image processing & machine learning algorithms we identify the type of fruit and its quality. An audio acknowledgment is given about the identification of the type of fruit while processing the fruit for packaging. In further enhancement we can develop a robot which can be used to separate the raw and ripe fruits with the help of detection algorithm used in this project.