Tile detection apparatus and detection method based on RGB-D camera

Abstract

The invention discloses a tile detection apparatus and detection method based on an RGB-D camera. The tile detection apparatus includes a frame, a conveyor belt conveying mechanism, a support, an RGB-D camera, light sources, a microcomputer and a liquid crystal display screen, wherein the support is displayed in a door type structure, and stretches over and is arranged on the conveyor belt conveying mechanism; the RGB-D camera and the light sources are fixed on the support; the conveyor belt conveying mechanism, the RGB-D camera and the light sources are controlled by the microcomputer; and the liquid crystal display screen is connected with the microcomputer. The detection method by means of the tile detection apparatus is high in detection efficiency, and can complete detection of surface flaws, tile size and flatness of tiles at the same time, wherein the detection process and the detection result accord with the related standard demand and quick and high accuracy detection work oftiles with different size can be realized. Besides, as non-contact detection is utilized during the detection process, the tile detection apparatus and detection method based on the RGB-D camera haveno harm to tiles.

Description

technical field [0001] The invention relates to the technical field of ceramic tile production, in particular to a tile detection device and detection method based on an RGB-D camera. Background technique [0002] The factors that affect the quality of tiles mainly include size, angle and surface defects chipping and cracks. On the one hand, the traditional contact measurement technology restricts the production efficiency and processing accuracy of ceramic tiles; affected product quality. If the fast speed, high precision, and online automatic detection cannot be realized, the production efficiency of the enterprise will be reduced, and even the economic benefit of the enterprise will be directly affected. At present, many ceramic enterprises mainly use manual inspection to inspect ceramic tiles. Workers use existing standard measurement tools to inspect the size, flatness, and appearance defects of tiles. The inspection efficiency is relatively low, and it is difficult t...

AI Technical Summary

Problems solved by technology

However, there are some problems and deficiencies in these technologies, and the detection function is relatively simple.

[0004] For example, the tile size and defect detection patent, such as the Chinese invention patent application number 201410017941.6, discloses a sliding tile surface flatness detection device and its method, which uses three laser rangefinders, and the tiles pass through the laser rangefinder Finally, the distance deviation is obtained, and the flatness of the tile surface is calculated from it, but this method cannot solve the problem of unstable measurement results caused by the jumping and displacement of the ceramic sheet during the movement process

Another example is the Chinese invention patent with application number 20150646060.5, which discloses a method and device for detecting the flatness of ceramic tiles. It uses several laser displacement sensors moving along linear guide rails to measure the distance between the tiles and the sensors, and calculates the flatness of the tiles. The method requires the ceramic piece to stop moving and perform detection, which takes a long time to detect and is inefficient

Another example is the Chinese utility model patent with application number 201520647768.8, which discloses a tile flatness detection device based on machine vision, which uses a structured light source and an area array camera to determine the flatness of the tile by detecting the shape of the structured light reflected on the tile It can perform dynamic detection during the moving process, but it can only detect the surface deformation in the horizontal direction of the plane (X-axis direction), and cannot detect the surface deformation in the longitudinal direction of the plane (Y-axis direction).

[0005] The above-mentioned patented technologies are limited to a certain type of measurement, and cannot simultaneously detect size, flaws and flatness

Images