External Quality Inspection Of Fruits Based On Multispectral Imaging

In the process of non-destructive quality inspection and automatic grading of fruits with machine vision and image processing technology, the stem and calyx were often confused with defects such as bruise and rot, and then was incorrectly distinguished and misclassified. The inspection methods of defects such as bruise and rot were studied, and the identification methods of stem and calyx were presented. The main research contents and results were as follows:1 The research advancements and achievements in the area of non-destructive quality inspection of fruits and automatic grading with machine vision and multispectral imaging were reviewed, and the existing problems were put forward.2 Multispectral imaging system suitable for this research was set up. This system is composed of a lighting chamber, eight fluorescent lamps (TL514W/840HE), CCD camera (HUBBA- III), filter table, filters, frame grabber (Meteor II/MC), and a computer (ADVANTECH INDUSTRIAL COMPUTER 610).3 Spectral reflectance experiments of apples were performed using FieldSpec HandHeld Spectrometer (Analytical Spectral Devices). The reflectance spectra of different apple tissues such as the sound, bruise, rot, stem and calyx were got through the experiments. The waveband was obtained to distinguish different apple tissues, and then the suitable filters were chosen to improve the precision of grading.4 B G R and I were used to denote the images captured at waveband of 480nm ± 10nm 530nm± 10nm 630nm± 10nm and 830nm± 10nm separately. And B G R and I were used as each component of RGB color model based on the RGB color model theory. RGB RGI GBI and RGBI, these four multispectral images were obtained when the components were added. Inspection results of defects such as bruise and rot show: the inspection precision of the sound is...