CN102788802A - A multi-camera workpiece quality detection method - Google Patents

Abstract

The invention provides a multi-camera workpiece quality detection method, wherein an applied detection mechanism comprises a plurality of cameras, a measurement and control system, a workpiece and a workbench, and the method is characterized by comprising the following steps of: a camera is mounted in the detection mechanism; detecting that a workpiece in the mechanism enters a camera shooting visual field; a step of acquiring a workpiece image by a camera in a detection mechanism; a camera in the detection mechanism transmits the workpiece image to a measurement and control system; and processing the received workpiece image by a measurement and control system in the detection mechanism. The invention has the advantages of realizing multi-angle detection and multi-view detection in workpiece quality detection, not only being suitable for general workpiece quality detection, but also solving the workpiece quality detection requirements of part of special materials and larger overall dimension, improving the work efficiency and detection precision of workpiece quality detection and having wide application.

Description

A multi-camera workpiece quality detection method

technical field

The invention relates to a workpiece quality detection method, in particular to a non-contact multi-camera workpiece quality detection method.

Background technique

With the advancement of science and technology and the development of society, people put forward higher requirements for the quality and testing methods of mechanical and electronic products. The quality of workpieces includes dimensions, shapes, materials and surface conditions. The detection tools in the traditional detection method of workpiece size mainly include ruler, vernier caliper and feeler gauge, etc., but in some specific occasions, such as the detection of small size, surface contour and hole distance, etc., it has become a difficult problem that cannot be realized by traditional detection methods. It is difficult to resolve the contradiction between precision and speed. Therefore, it is of great significance to explore new detection methods.

With the development of photoelectric technology, computer technology and precision machinery technology, image measurement technology emerges as the times require. Image measurement uses image probes to collect images of workpieces, and extracts coordinate points on the surface of workpieces with various complex shapes through digital image processing technology. Use coordinate point transformation and data processing technology to transform into various geometric elements in the coordinate measurement space, so as to calculate the actual size, shape, surface condition and mutual position relationship of the measured workpiece. After continuous development, the application range of image measurement technology has been continuously expanded, and it can precisely measure various complex workpiece contours and surface shapes. Now, the application of image measurement technology has gradually entered the electronics, machinery, instruments, clocks, light industry, national defense and military industry and In industries such as aerospace and aviation, it has become a commonly used measurement method in laboratories, measurement rooms and production workshops of institutions of higher learning, research institutes, and measurement technology institutions.

However, the image probe used in the general image measurement method is a fixed single lens, and there are the following problems in the detection: for workpieces with large external dimensions, when the external dimensions exceed the shooting range of the lens, it is necessary to move the workpiece and take multiple shots to complete the workpiece. The collection of complete images is inefficient; for workpieces with special materials, the surface flaws must be observed from multiple angles, such as backlight modules, foreign objects and dander on the surface are relatively clear when viewed vertically at 90 degrees, while Scratches and dirt can only be clearly displayed when observed at about 30 degrees. It can be seen that for the detection of workpiece surface defects of some special materials, it cannot be detected when using a fixed single lens, and the detection accuracy is poor. Therefore, the existing image measurement technology is an important means of workpiece quality inspection, and there is a need for multi-angle inspection and multi-view inspection.

Contents of the invention

In order to realize the multi-angle detection and multi-view detection of image measurement technology, the present invention provides a multi-camera workpiece quality detection method, the detection mechanism used includes multiple cameras, measurement and control systems, workpieces and workbenches, specifically including the following steps:

The steps of camera installation in the inspection mechanism;

Steps in which the workpiece in the detection mechanism enters the field of view of the camera;

Steps for the camera in the inspection mechanism to acquire images of the workpiece;

The step in which the camera in the inspection mechanism transmits the image of the workpiece to the measurement and control system;

A step in which the measurement and control system in the detection mechanism processes the received image of the workpiece.

The number of cameras in the detection mechanism is not less than two, which are installed at a suitable position above the workbench. The cameras in the detection mechanism are connected to the measurement and control system through cables. Multiple cameras in the detection mechanism take pictures of the workpiece at the same time, and each camera only acquires one image of the workpiece. When multiple cameras in the detection mechanism are installed at different angles, multiple cameras are directed at the same area of the workpiece at different angles at the same time, and each camera in the detection mechanism acquires images of the same area of the workpiece at different angles, realizing multi-angle detection, suitable for Workpiece defect detection of some special materials; when multiple camera imaging surfaces are installed in parallel in the detection mechanism, multiple cameras are directed to different areas of the workpiece, each camera in the detection mechanism obtains a partial image of the workpiece, and all partial images of the workpiece are synthesized in sequence The complete image realizes multi-view detection, which is suitable for workpiece quality inspection when the outer dimension of the workpiece exceeds the single lens shooting field of view. Therefore, the common determinants of the number of cameras in the detection mechanism, the installation position and the installation angle are the external dimensions of the workpiece and the material properties of the workpiece itself.

The workpiece in the detection mechanism is on the workbench. When the workbench in the detection mechanism is fixed, the workpiece is directly put into the camera field of view on the workbench by the operator. When the workbench in the detection mechanism is an assembly line, the workpiece enters the work through the assembly line. The camera on the stage captures the field of view.

The measurement and control system in the detection mechanism performs image processing and image analysis on the received workpiece image data according to the corresponding project requirements, and obtains the workpiece quality detection results.

The beneficial effects of the present invention are:

1. The invention can realize non-contact detection, has a simple mechanism and no moving parts, and has a high response speed.

2. The present invention can realize multi-angle detection and multi-view detection in workpiece quality detection, not only applicable to general workpiece quality detection, but also solves the quality detection needs of some special materials and large-scale workpieces, and improves the efficiency of workpiece quality detection Work efficiency and detection accuracy are more widely used.

3. The present invention can complete the entire workpiece quality detection task after the workpiece flows through the detection mechanism once, which greatly improves the workpiece quality detection efficiency.

Description of drawings

FIG. 1 is a schematic diagram of the present invention applied to defect detection of a backlight module.

Fig. 2 is a schematic diagram of the application of the present invention to the detection of workpieces with large external dimensions.

Fig. 3 is a flow chart of the application of the present invention to workpiece defect detection.

The main structures in Figure 1 and Figure 2 are: 1-camera, 2-assembly line, 3-backlight module, 4-measurement and control system, 5-workpiece.

Detailed ways

The present invention will be described in further detail below in conjunction with the accompanying drawings.

Example 1

This embodiment is the flaw detection of the backlight module in the production process of the automatic assembly line. Referring to FIG. 1 and FIG. The measurement and control system 4 is connected by cables, and the backlight module 3 is on the assembly line 2 . One camera in camera 1 is installed at 30 degrees to the plane where assembly line 2 is located, and the other camera is installed at 90 degrees to the plane where assembly line 2 is located.

The defect detection process of the backlight module 3 is as follows: the backlight module 3 gradually approaches the shooting field of view of the camera 1 through the transmission of the assembly line 2, and the backlight module 3 enters the shooting field of the camera 1; The camera shoots and acquires two images of the backlight module 3 at the same time, one of which is the image of the backlight module when it is observed at 30 degrees. The image when the module is observed at 90 degrees, which shows the image status of the backlight module 3 foreign matter and dander; the camera 1 transmits the two images of the backlight module 3 acquired to the measurement and control system 4 through the cable; the measurement and control system 4 After receiving the two images of the backlight module 3 transmitted by the camera 1, the measurement and control system 4 uses software to analyze the data of the foreign objects, dander, scratches and stolen goods on the two images, and compares them with the project specifications: when the two images When the defect data of only one of the backlight module 3 images exceeds the specification, the measurement and control system 4 determines that the quality of the backlight module 3 is unqualified; when all the defect data in the two images of the backlight module 3 meet the specifications, the measurement and control system 4 determines that the backlight The quality of module 3 is acceptable.

Example 2

This embodiment is for the detection of workpiece defects with large dimensions in the production process of the automated assembly line. Referring to Figures 2 and 3, the imaging surfaces of the three cameras in the camera 1 are installed in parallel above the assembly line 2 and face different areas of the workpiece 5. The cameras 1 and 5 are The measurement and control system 4 is connected by cables, and the workpiece 5 is on the assembly line 2 .

The defect detection process of the workpiece 5 is as follows: the workpiece 5 gradually approaches the shooting field of view of the camera 1 through the transmission of the assembly line 2, and the workpiece 5 enters the shooting field of view of the camera 1; the camera 1 takes pictures of the workpiece 5, and the three cameras shoot at the same time and obtain the partial image of the workpiece 5 respectively. The three partial images of the workpiece 5 captured by the camera 1 are sequentially synthesized into a complete image of the workpiece 5; the camera 1 transmits the three partial images of the workpiece 5 to the measurement and control system 4 through the cable; the measurement and control system 4 receives the three partial images of the workpiece 5 transmitted by the camera 1 Image, the measurement and control system 4 uses software to analyze the data of the defects on the three partial images of the workpiece 5, and compares them with the project specification requirements: when the defect data of only one of the three partial images of the workpiece 5 exceeds the specification, the measurement and control system 4. Determine that the quality of the workpiece 5 is unqualified; when all the defect data in the three partial images of the workpiece 5 meet the specifications, the measurement and control system 4 determines that the quality of the workpiece 5 is qualified.

Claims (5)

1. the workpiece quality detection method of a polyphaser, the testing agency of its application comprises a plurality of cameras, TT&C system, workpiece and worktable, it is characterized in that may further comprise the steps:

The step that camera is installed in the testing agency;

Workpiece gets into the step that camera is taken the visual field in the testing agency;

Camera obtains the step of workpiece image in the testing agency;

Camera sends the workpiece image to the step of TT&C system in the testing agency;

TT&C system carries out processed steps to the workpiece image that receives in the testing agency.

2. the step that camera is installed in the testing agency as claimed in claim 1 is characterized in that:

Camera is installed in worktable top place in the testing agency;

The quantity of camera is no less than two in the testing agency;

The quantity of camera, installation site and setting angle codeterminants are the physical dimension and workpiece self material behavior of workpiece in the testing agency.

3. workpiece gets into the step that camera is taken the visual field in the testing agency as claimed in claim 1, it is characterized in that:

Workpiece is on the worktable in the testing agency;

When worktable was fixed in the testing agency, workpiece was taken the visual field by the camera that operating personnel directly put on the worktable;

When worktable was streamline in the testing agency, workpiece was taken the visual field through the camera that the streamline transmission gets on the worktable.

4. camera obtains the step of workpiece image in the testing agency as claimed in claim 1, it is characterized in that:

Camera in the testing agency is taken workpiece simultaneously, and each camera only obtains a width of cloth workpiece image;

Simultaneously when the workpiece same area, each camera in the testing agency obtains the image of workpiece same area different angles to a plurality of cameras separately with different angle in the testing agency;

The parallel placement of camera imaging face and when the workpiece zones of different in the testing agency, each camera in the testing agency obtains the workpiece local image separately, and all workpiece local images are the complete image of synthetic workpiece according to the order of sequence.

5. TT&C system carries out processed steps to the workpiece image that receives in the testing agency as claimed in claim 1, it is characterized in that:

TT&C system in the testing agency requires the workpiece image data that receives is carried out Flame Image Process and graphical analysis according to respective item, draws the workpiece quality testing result.

Images