A method of self-adaptive retrieving of array material frame based on machine vision

Abstract

The invention discloses a self-adaptive material taking method based on machine vision and used for an array type material box. The self-adaptive material taking method is used for taking material from the material box in which materials are placed in an array mode. The self-adaptive material taking method comprises the steps that the standard placing position for placing the material box is set, the standard photographing position of the first material is determined, and the standard image coordinates of key characteristics of all the materials are marked visually; the image of the first material is obtained in the standard photographing position, a complete material image cannot be obtained if the material box position deviates from the standard position, the photographing position of the first material is adjusted in a self-adaptive mode according to local image characteristics of an incomplete material image so that the complete material image can be obtained, the actual image key characteristic position is obtained accordingly, accurate visual grabbing and positioning are achieved, and the current position of a movement mechanism is set to be the actual photographing position of the first material; and the photographing position of the current material is adjusted in a self-adaptive mode in real time according to the photographing position of the prior adjacent material and the key characteristic position of the image of the prior adjacent material, so that a complete material image of the current material is obtained, and accordingly accurate visual grabbing and positioning are achieved.

Description

technical field [0001] The invention belongs to the technical field of production automation, and in particular relates to a machine vision-based self-adaptive retrieving method for an array type material frame. Background technique [0002] For most modern industrial automation production lines, there are ubiquitous automatic pick-and-place material devices, such as manipulators, modules, etc. In order to complete the "grab-and-place" action, it is particularly necessary to obtain material positioning information. In order to achieve higher positioning and grasping accuracy, machine vision is usually used for positioning. In order to improve the efficiency of feeding, the general industrial site adopts the array-type material frame that can be loaded in batches at one time, and the placement position of the material is basically fixed according to a certain rule, which is convenient for visual positioning; however, considering the actual situation of the industrial site Si...

AI Technical Summary

Problems solved by technology

In order to improve the efficiency of feeding, the general industrial site adopts the array-type material frame that can be loaded in batches at one time, and the placement position of the material is basically fixed according to a certain rule, which is convenient for visual positioning; however, considering the actual situation of the industrial site Situation: (1) The general material frame placement point is in a fixed area, so there will be a certain deviation in the position every time the material frame is placed manually or the AGV trolley feeds; (2) The space for placing products in the material frame will exist There is a certain margin, so it cannot be guaranteed that each product can be placed in a very fixed position, and there is also a certain deviation

[0003] Generally, in order to accurately grasp the material, the grasping device based on machine vision must obtain clear material image features, so the material image features obtained within a certain camera lens field of view should be as large as possible; if As mentioned above in the actual situation of the industrial site, there is a certain deviation in the position of the material frame and the material, which may lead to the inability to take a complete material image within the field of view of the lens, so that the image features of key parts cannot be obtained, and it is impossible to capture Take positioning; this is easy to produce missed grasping, reducing the reliability of grasping, and the efficiency of retrieving materials is low

Images