CN106052793A - Machine vision based liquid level sub-quality product marking method - Google Patents

Abstract

The invention relates to a method for marking unqualified liquid level products based on machine vision. position, take the image of the liquid level; perform binary processing on the preprocessed image; use the geometric matching method to match the bottle cap, and detect the position of the bottle cap; establish a coordinate system; set the ROI of the liquid level qualified area ; Use the geometric matching method to match the liquid level line again; judge the unqualified glass bottles. The algorithm used in the invention is simple, has few steps, realizes rapid detection of images, has high detection efficiency, is less affected by the environment, and has high accuracy.

Description

A method of marking unqualified liquid level products based on machine vision

technical field

The invention relates to a glass bottle liquid level detection unqualified marking method, which belongs to the field of machine vision.

Background technique

In industrial production lines such as beverages and beer, most of them use manual visual inspection to check whether the bottles are qualified.

Especially in filling production, whether the liquid level of bottled beverages is consistent and whether the height is uniform has a huge impact on the enterprise in the market. However, manual inspection has many defects such as slow speed, low efficiency, unstable inspection quality,

Omissions or false detections often occur, resulting in unstable product quality. Once defective products are produced, it will cause waste. Aiming at the above-mentioned defects of manual detection, the research on bottle cap detection based on machine vision and the establishment of a detection system based on the platform of optical imaging, image acquisition, digital image processing and analysis not only have theoretical basis, but also have great economic value , can ensure to reduce and eliminate the occurrence of the above phenomena to the greatest extent, and improve product quality.

Machine vision detection has been applied in many detection fields, and there are many application precedents in liquid level detection.

However, most of them obtain the actual height value of the liquid level by various means, such as the patent "a liquid level detection device and method after beverage filling" (Patent No.: CN101858768A). There is a large error in the liquid level height. In an industrial production line, at least 5 bottles must be detected per second, and if unqualified, it must be marked or rejected. This has high requirements for the real-time performance of the detection system and software processing.

Contents of the invention

The purpose of the present invention is to overcome the shortcomings of manual detection and the limitations of liquid level detection, and provide an efficient method for marking glass bottle liquid level unqualified products based on machine vision.

Technical scheme of the present invention is as follows:

A method for marking unqualified liquid level products based on machine vision, comprising the following steps:

1) Place a qualified glass bottle at the station to be tested, use the LED white backlight to illuminate the liquid level position of the glass bottle to be tested, the camera collects the liquid level information of the bottle cap, adjust the camera and light source, so that the liquid level of the glass bottle collected is in line with the level below The beverage is separated to obtain a clear liquid level line for subsequent matching processing;

2) The glass bottle to be tested arrives at the shooting station through the conveyor belt, the photoelectric sensor receives the signal and triggers the camera to take the image of the liquid level to be tested, and performs median filter preprocessing on the image to remove noise interference and enhance the detection target information;

3) Perform binary processing on the preprocessed image, adjust the appropriate threshold according to the actual environment, so that the binarized image can effectively separate the liquid level from the background;

4) Use the geometric matching method to match the bottle cap: intercept the bottle cap side view of the glass bottle image to be tested, set the geometric center point of the bottle cap as the matching base point, extract the edge of the bottle cap as a template for geometric matching, and detect the bottle cap through geometric matching the position of the cover;

5) establish a coordinate system according to the matching base point in step 4 as the origin;

6) With the origin of the coordinate system established in step 5 as a reference point, according to the principle that the qualified liquid level region does not change relative to the geometric center point of the bottle cap, the region of interest ROI of the qualified liquid level region is set;

7) Use the geometric matching method to match the liquid level line again: intercept a section of the liquid level line, extract its edge as a template for geometric matching, and perform edge detection. If the matching is successful, it is judged as a qualified product. If there is no liquid level in the qualified area, it is judged as unqualified;

8) For glass bottles that are judged to be unqualified, at the downstream station of the station, output signals to perform marking operations.

The beneficial effects of the present invention are as follows: firstly, the algorithm used in the method is simple, the steps are few, the rapid detection of the image is realized, and the detection efficiency is high; secondly, the method is less affected by the environment and has a high accuracy rate; Good; the detection method is novel and has a wide range of applications, and is applicable to the detection of industrial glass bottled products; this method can be added to the existing production line to realize on-line detection during production. The invention is suitable for the marking work of products with unqualified liquid levels in glass bottles after filling in automatic production.

Description of drawings

Fig. 1 is a schematic diagram of a marking device for unqualified beverage bottle liquid level detection based on machine vision used in the present invention.

Fig. 2 is based on machine vision of the present invention to glass bottle liquid level detection method flow chart

Figure 3 (a) and (b) are the original liquid level image to be tested and the liquid level image collected in step 1 of the present invention respectively.

Fig. 4 (a), (b) and (c) are respectively the judgment result diagrams of the qualified liquid level, exceeding qualified liquid level and lower than qualified liquid level detected by the present invention.

detailed description

The present invention will be further described below in conjunction with the accompanying drawings and specific examples.

Referring to Figure 1, a system device for marking unqualified beverage bottle liquid level based on machine vision is divided into 6 modules in terms of function, namely conveyor belt 1, image acquisition module 2, photoelectric trigger 3, lighting module 4, image processing and identification 5 and output control means 6 . The above-mentioned image acquisition module 2 uses a 1.2 million-pixel CCD camera, the lighting module 4 uses an LED backlight board, and the image processing and recognition module 5 uses an industrial computer. The output control device consists of a nozzle controlled by a solenoid valve. The conveyor belt 1 transports the bottle to be tested to the photographing station, and the photoelectric trigger device 3 is used to detect whether the bottle arrives at the photographing position, so as to generate a pulse signal and deliver it to the industrial computer 5, and the industrial computer sends a signal to trigger the camera 2 to take pictures; the LED backlight is The camera provides lighting to improve the image quality when taking pictures. The CCD camera is located directly above the conveyor belt 1 and takes pictures of the liquid level of the glass bottle in front of the backlight plate. The captured images are sent to the industrial computer for processing; the output control device is connected to the industrial computer. When an unqualified product is detected, a control signal is output to control the nozzle 6 to spray paint to mark it.

Referring to FIG. 2 , it is a flow chart of the machine vision-based method for marking unqualified beverage bottle filling level products according to the present invention. Proceed as follows:

(1) Place any qualified bottle on the station to be tested, use LED white backlight to illuminate the liquid level position of the beverage bottle to be tested, use a high-resolution camera to collect the liquid level information of the bottle cap, adjust the camera and light source, so that the collected beverage bottle The liquid level is separated from the beverage below, resulting in a clear liquid level line.

(2) Start the conveyor belt, the bottle to be tested arrives at the shooting position along with the conveyor belt, trigger the camera to acquire the image of the bottle cap to be tested, and transmit it to the image processing system of the industrial computer through the channel.

(3) Median filter preprocessing is performed on the image, and the liquid level information of the image is significantly enhanced after processing. The processed image is binarized, and the appropriate threshold is adjusted according to the gray histogram to increase the image recognition rate, so that the binarized image can effectively separate the liquid level from the background.

(4) Use the geometric matching method to match the position of the bottle cap, intercept the bottle cap side view of the beverage bottle image to be tested, extract the edge of the bottle cap as a template for geometric matching, and set the geometric center point of the bottle cap as the matching base point. Set the matching number to 1, the matching angle range from -10 to +10, the matching algorithm is edge detection, the minimum matching score is 500, and the exact matching score is 1000. Establish a coordinate system with the matching base point as the origin

(5) Taking the origin as a reference point, set a rectangular area with a length of 360 and a width of 250 as the liquid level qualified area and set it as the ROI (Region Of Interest).

(6) Use the geometric matching method to match the liquid level line again, intercept a relatively standard part of the liquid level line, and extract its edge as a template for geometric matching. Also set the matching number to 1, and the matching angle range is -10—+ 10. The matching algorithm is edge detection. Through geometric matching, if there is a liquid level in the qualified area, the matching is successful, and it is judged as a qualified product. If there is no liquid level in the qualified area, it is judged as unqualified.

(7) When the unqualified liquid level is detected, the control signal is output through the serial I/O, the solenoid valve is opened, the nozzle of the marking unit sprays red pigment to mark the unqualified bottle, and then go to step 2 to continue to detect the next bottle cap.

As shown in Figure 3, the image on the left is the original image captured by the camera, and the image on the right is the image collected after adjusting the exposure rate of the camera and the light source. This adjusted image makes subsequent processing and judgment easier.

See Figure 4. The above figure shows the judgment result of the qualified liquid level. It can be seen from the figure that the qualified detection area set with the geometric center of the bottle cap as the reference point matches the liquid level line. The picture in the middle shows the test results when the liquid level is higher than the qualified area, and the picture below shows the test results when the liquid level is lower than the qualified area. It can be seen that when the filling liquid level is too high or too low, it will be judged as a substandard product .

Claims (1)

1. a labeling method for liquid level defective work based on machine vision, comprises the following steps:

1) certain grade glass bottle is placed on station to be measured, illuminates vial liquid level position to be measured, camera with LED white backlight plate Gather bottle cap liquid level information, regulation camera and light source, make the vial liquid level of collection separate with beverage below, obtain one clearly Clear liquid level line, in order to follow-up matching treatment；

2) vial to be measured arrives shooting station by conveyer belt, and photoelectric sensor receives signal triggering camera shooting and treats its liquid level Image, carries out medium filtering pretreatment to image, to remove noise jamming and to strengthen detection target information；

3) pretreated image is carried out two-value process, adjust suitable threshold value according to actual environment, make the figure after binaryzation As effectively liquid level and background being separated；

4) geometric match method coupling bottle cap is used: intercept the bottle cap side view of vial image to be measured, set in bottle cap geometry Heart point is coupling basic point, and the edge of extraction bottle cap, as the template of geometric match, detects the position of bottle cap by geometric match；

5) coordinate system is set up according to the coupling basic point of step 4 as initial point；

6) coordinate origin set up with step 5 is as a reference point, according to the qualified region of liquid level relative to bottle cap geometric center point The principle that position does not changes, sets the region of interest ROI in the qualified region of liquid level；

7) geometric match method matching fluid bit line is reused: intercept one section of liquid level line, extract its edge as geometric match Template, carry out rim detection, if there is liquid level in the region of interest ROI by the qualified region of geometric match liquid level, It is made into merit, it is determined that for qualified products, if there is no liquid level in qualified region, be then judged to defective；

8) for being judged to underproof vial, in the downstream process of described station, output signal execution flag operates.