Fruit recognition using Statistical and Features extraction by PCA

Abstract

Fruits are an integral part of human diet since they are a vital source of minerals, vitamins, fiber, and phytonutrients. Fruits are rich in potassium, fiber, and vitamin C yet low in fat, sodium, and calories. A diet high in fruit can help us avoid diseases including cancer, diabetes, heart disease, and others. Without professional dietitian guidance, a method that quickly reveals how many calories or fruit they are consuming can be helpful in maintaining health. The use of image processing methods is expanding across all academic fields, including food science and agriculture. The identification of plant fruits and the extraction of their features are the first topics covered in this essay because they are essential to agriculture. The goal is to use the results of Principal Component Analysis (PCA) to build an accurate, efficient, and reliable framework. Fruit detecting software could simplify human labor. Based on color and shape characteristics, several fruit recognition methods have been developed. However, the color and shape values of several fruit photos could be comparable or even the same. As a result, utilizing PCA feature extraction analysis methods to identify and distinguish fruit photos is still not strong and effective enough to boost recognition accuracy. In this paper, a fruit recognition algorithm based on Principal Component Analysis (PCA) is proposed. The establishment of a database of fruit photos with 6 distinct categories and 36 photographs is the second topic covered in this essay. In this study, a PCA classifier is used to implement the system, and the proposed system's classification accuracy is 75%. KEY WORDS: Fruit, recognition, Feature extraction, Fruit Classification, Principal Component Analysis (PCA).