Automatic detection device and method for spike-shaped fruit based on monocular multi-view imaging

Abstract

The invention provides an automatic spikelike fruit detection device based on monocular multi-view imaging. The automatic spikelike fruit detection device comprises a transmission module, a plane mirror group, an imaging component, a data processor, a display and a sorting device, wherein the transmission module comprises a conveyer belt, a chain transmission motor and suspension clamping mechanisms; a plurality of groups of suspension clamping mechanisms are arranged at the lower part of the conveyer belt; a weighing module is mounted above the conveyer belt; the plane mirror group consists of two pieces of plane mirrors which are of the same size and are vertically placed; front surfaces of the two plane mirrors are arranged opposite to each other and are plated with aluminum; light can be reflected by the surfaces plated with aluminum; the imaging component comprises a CCD camera and a polarizer positioned at the front end of the CCD camera; the weighing module and the output end of the CCD camera are connected with the input end of the data processor; the output end of the data processor is connected with the display and the sorting device. The automatic spikelike fruit detection device is simple in structure and relatively low in cost, and automatic detection on spikelike fruits can be achieved.

Description

technical field [0001] The invention belongs to the field of non-destructive detection of spike-shaped agricultural products, and in particular relates to an automatic detection device and method for spike-shaped fruits based on monocular multi-view imaging. Background technique [0002] Post-harvest commercialization of fruits is very important. Fruits are graded and classified according to quality before they enter the consumer market, and prices can be assessed according to quality, which not only improves their market competitiveness and added value of commodities, but also satisfies different levels of consumer demand. At present, postharvest fruit commercialization can be divided into internal quality and external quality. Computer vision technology has been successfully applied to the appearance quality inspection of food and agricultural products with its advantages of fast, non-destructive, and easy online inspection. For example, the online inspection equipment pro...

AI Technical Summary

Problems solved by technology

However, the contours captured by the two cameras are actually the same projected contour, and the contour edge extraction of the target object is not sufficient.

There are public documents that place plane mirrors on both sides of the conveyor belt to obtain the side information of the fruit, but this method is blocked by the conveyor belt and its rolling action is not suitable for spike-shaped fruits and cannot reflect all the information of the fruit

Looking at the shortcomings of the existing computer vision technology for fruit appearance detection, especially the lack of effective and simple detection methods for spike-shaped fruits such as grapes, resulting in a low level of commercialization of spike-shaped fruits

Images