Automatic Bruise Detection in Apple Using Thermal Camera

New safe and fast methods for grading of fruits have important place in agricultural economy. Inconsistencies associated with manual grading decrease when automated grading systems are used. Thus, error rate and costs decreases while speed increases. Large scale utilization of automatic classification system for this purpose demands on a robust color classification under color saturation, variations of environment lighting and light reflections. As known size, shape, color and tissue are basic criteria in the classification process. In this study, automatic apple grading by size and color using thermal camera and computerized image processing technique is proposed. It is very tedious and hectic job to monitor fruit bruise manually and time consuming process so BDS (Bruise Detection System) is used for the detection of fruit diseases. BDS involves some basic steps of image processing system and its classification using k-means clustering method. The use of ANN (artificial neural network) methods for classification of bruises in fruits like support vector machine is efficiently used. This method is used accurately to identify and classify various fruit bruises.