Method for detecting flaw on surface of fruit

Abstract

The invention provides a method for detecting a flaw on the surface of a fruit by using image processing in the technical field of image processing. The method comprises the following steps of: (1) obtaining a RGB (red, green and blue) image and an NIR (near infrared reflection) image; (2) removing the backgrounds of the RGB image and the NIR image to respectively obtain an RGB foreground image and an NIR foreground image only containing a fruit image; (3) transforming the RGB foreground image into a YCrCb color space image from an RGB color space image, and marking a new image as a new space image; (4) carrying out specific value computation on a Y component image of the new space image and the NIR foreground image to obtain a specific value image; and (5) extracting a fruit flaw image from the specific value image. According to the method, the flaw on the surface of the fruit can be stably, exactly and quickly detected, and the influence caused by the shape and the size of the fruit can be effectively avoided.

Description

technical field [0001] The invention relates to a method in the technical field of image processing, in particular to a method for detecting fruit surface defects by using image processing. Background technique [0002] Fruit is one of the popular agricultural products loved by the people. According to reports, since 1990, the output value of major fruits in the world has shown an increasing trend. The surface defect of fruit is the most direct reflection of its quality, so surface defect is an important factor to determine the price of fruit. Detection and sorting of fruit surface defects is a necessary link before fruit sales, processing and storage. At present, machine vision is widely used in the detection of the external quality of fruits. The camera collects the image of the fruit surface, and then transmits it to the computer. The size, color, defect and other quality characteristics of the fruit are extracted through image processing technology, and then according t...

AI Technical Summary

Problems solved by technology

The research results show that this method has a high defect recognition rate. However, due to the complexity of the algorithm, it is not conducive to the online detection of fruit external quality.

At the same time, fruit is a spherical organism, so fruit can be regarded as a Lambertian body. According to the principle of Lambertian reflection, the brightness of any point on the sphere is the normal vector of the point and the angle between the point and the light source. Cosine is proportional to, that is, I D = I L ×COSθ, where I D is the reflected light intensity, I L is the incident light intensity, so in order to improve the imaging quality when collecting images, it is generally necessary to add an additional light source, so the Lambertian phenomenon will occur on the surface of the fruit, so the collected image forms a distribution with bright edges and dark edges in the middle, and the normal area of ​​the edge of the fruit is easily misunderstood. identified as defective areas

[0003] After searching the prior art, it was found that Chinese Patent Document No. CN102788806A, published on 2012-11-21, records a method for detecting fruit surface defects based on spherical brightness transformation, using fruit RGB images and NIR images to compare and calculate The shape and size of the defect of the fruit, but the defect of this technology is that the fruit is not strictly spherical. This patent uses the maximum width Y of the circumscribed rectangle of the binary image to replace the fruit diameter, half of the maximum width, that is, Y / 2 as The termination condition of the number of iterations, so this method has certain limitations for elliptical fruits with less circularity; at the same time, the degree of Lambert phenomenon near the major axis and the short axis of elliptical fruits is very different. The patent is based on the binary value The number of pixel points M of the edge pixels of the image, the sum of the brightness of all points on the edge of the R component image is divided by the number of pixel points M to obtain the average brightness, but for the same number of iterations near the long axis and short axis of the elliptical fruit It is not very reasonable to average the brightness of pixels affected by different lighting, and it will also have a certain impact on defect recognition.

Images