Fruit epidermis micro-damage detection method based on X-ray and contrast enhancer

Abstract

The invention relates to a fruit epidermis micro-damage detection method based on X-rays and a radiography enhancer. The method comprises the following steps: step 1, soaking and cleaning the epidermis of a fruit to be detected, and adding a potassium iodide radiography enhancer into a cleaning solution; 2, imaging the fruits to be detected for multiple times through the X-ray imaging device to obtain X-ray images of the fruits to be detected; and step 3, inputting the X-ray image of the to-be-detected fruit into the fruit detection deep learning model, and identifying damage features of the X-ray image of the to-be-detected fruit and outputting an identification result by the fruit detection deep learning model. The damaged fruits can be effectively and quickly screened out, and the damaged fruits are prevented from being affected by diseases to healthy fruits.

Description

technical field [0001] The invention relates to the technical field of agricultural product detection, in particular to a method for detecting micro-damages on fruit epidermis based on X-rays and contrast enhancers. Background technique [0002] Micro-damage to the fruit epidermis will bring huge risks to the post-harvest storage stage of the fruit. Micro-damage fruits are very easy to be infected with pathogens in the air and get sick (such as penicillium, green mold, sour rot, etc.), diseased fruits It will infect adjacent healthy fruits, cause economic losses to fruit farmers, and affect the eating quality of other fruits. [0003] The traditional identification of micro-damages on fruit epidermis mainly relies on manual sorting, which is extremely difficult to identify with naked eyes, time-consuming, labor-intensive, inefficient, and labor-intensive. After the fruit is harvested, the collision and the pricking of the stems and branches are very easy to produce micro-da...

AI Technical Summary

Problems solved by technology

[0004] The difficulty of using visual technology to identify micro-damages on fruit epidermis is that the micro-damages are too small and the damage characteristics are not obvious enough

It is difficult to detect micro-damages on fruit epidermis based on traditional visual technology, and it is difficult to capture micro-damages on fruit epidermis with RGB cameras; there are also the following difficulties in detecting micro-damages on fruit epidermis only with direct digital X-ray photography technology: the fruit tissue is complex and the difference in tissue density is small. There is little difference in imaging gray scale, and there is noise interference in X-ray imaging of fruit

Images