Glass bottle manufacturing process that meets market demand
Glass bottle manufacturing process that meets market demand In the production process of glass bottles, various defective products will be produced, which will bring serious hidden troubles to the products. In order to improve the quality of glass bottles, glass bottle manufacturers now rely on manual checks to remove waste. However, the speed of manual inspections requires a lot of manpower, material resources and site resources, and the eyes are prone to fatigue and negligence. With the continuous development of computer technology, now it has gradually used computers instead of manual quality testing, so that testing standards can be set by the system, without subjective factors, continuous and accurate completion of the detection work. The main detection functions of the system include: Software Design The entire system software is developed using VC++6.0 and is divided into two parts: the server part completes the information synthesis, data storage, and subsequent statistical analysis of test results; the detection part mainly completes image acquisition and defect analysis. For the detection system, the core part is the defect detection algorithm part, and the system algorithm part is based on the DHImage library. The detection algorithm flow is shown in Figure 3. Due to the jitter that occurs when the glass is transported in the production line, the image of the glass bottle is tilted. In order to accurately measure the relevant dimensions, we must locate the glass bottle. The system first looks for the outline of the glass bottle, and then uses the method based on the contour registration to accurately position, which provides a good guarantee for subsequent dimension measurement. In order to reduce the shadow interference of the image edge and obtain the largest possible detection area, we added a search and detection area before the defect analysis. The main method is to automatically calculate the internal actual area according to the characteristics of the glass bottle, and at the same time, Defects (bubbles, cracks) are predicted. In the analysis of the BLOB (connected area block, may be a defect or interference), we fully consider the influence of uneven frost on glass bottles, non-uniform thickness of glass bottles, etc., and inconspicuous defects such as thin skin and bombing. In-depth analysis of features, some of which are shown in Figure 4 as an example of a defect image, develops a set of algorithm modules that specifically address defects in glass bottles. Figure 4 partial defect image Field test results During the system development stage, we used mechanical equipment such as conveyor belts and bottle conveyors in the laboratory to design and install a glass bottle automatic conveyor system. After one month of system testing in the laboratory, we The system is installed in stages in white bottles, white scale bottles, brown bottles, brown scale bottles, and vial production lines. The system's main monitoring interface is shown in Figure 5. In actual use, we improve the detection algorithm according to the characteristics of various types of glass bottles, and finally carry out optimization and integration. Now the detection system can flexibly detect different types of glass bottles according to the production needs according to different options. sample is available for pet rain coat,dog rain coat,cat rain coat and etc.
welcome to discuss the details,we will give the professional suggestion and sent the sample for check the quality.
Dongyang Yadi Pet Product Co , Ltd was founded in 2006 and covers an area of 6500 square meters . We are located in Zhejiang , the famous manufacturing city in China.We are specialized in developing and manufacturing many kinds of pet products. Our featured products are pet collar ,PET Leash ,pet harness, pet clothes,pet bandana,small pet products,ect.
Our company also o ffer all kinds of OEM/ODM services.And enjoy a good reputation among clients,we can provide you with the highest quality products and services.
In order to take up the market , our company focuses on introducing and cultivating talents as well as constantly developing new products.We always take the principle of [credibility , concentrate on work , quality supreme , innovation and development", with a heart of thanks giving to return customers and society.
Welcome to visit our company!
Pet Rain Coat,Pet Rain Coat For Sale,Dog Rain Coat,Cat Rain Coat Dongyang Yadi Pet Product Co.,Ltd , http://www.dyyadipet.com
Glass bottle quality inspection requirements The system is designed to detect defects in pharmaceutical glass bottles, including white bottles, brown bottles, and graduated glass bottles, as shown in Figure 1. 
Figure 1 Example of a glass bottle to be tested
A. Inspection of dimensions (including glass bottle height, outer diameter of glass body, outer diameter of glass bottle mouth, height of glass bottle mouth, etc.);
B \ Glass bottle appearance defects (including bubbles, impurities, folds, horizontal and vertical stripes, adhesions, stones, cracks, nicks, abrasions and obvious oil, fingerprints, etc.);
C. Defects in the bottom of the glass bottle can be photographed from the side (including uneven bottom, bottom barb, bottom, etc.);
D. Defects in the shoulders of glass bottles (including oblique shoulders, bottles, appearance defects similar to the bottle body, etc.);
E. The inspection content of the glass bottle mouth part is to detect defects such as notches, breaks, and round mouth defects in addition to the defect types of the glass bottle body.
Hardware design According to the requirement of detecting defect types, the system uses five high-speed cameras. When the glass bottles are moving on the conveyor belt, the system uses external triggering to accurately capture four-sided images and one glass bottle mouth image at fixed positions. Images are transferred to two high-performance processors for processing and analysis. The results are aggregated on a single server for unified control and display. The camera chooses Daheng DH-HV1310FM with a resolution of 1280*1024 and supports 8bit or 12bit output. The angle between the four side cameras is 90 degrees to ensure that there is no dead zone for the entire bottle. The glass bottle mouth part is processed by a single camera, and a bowl-shaped light source is used to take images.
According to the comprehensive analysis of the characteristics of the defects, it was found that for the outer defects of glass bottles, the lighting method using high-brightness backlights can well reflect various defects; for the surface defects of glass bottles, considering the characteristics of the production line and fully reflecting the surface of the glass bottle mouth Defects require the design of a bowl-shaped light source so that defects in the glass bottle mouth can be well detected without changing the existing production line equipment.
Since the type of glass bottle on the production line may be frequently changed according to demand (glass bottle height range 15-150mm), we take this factor into consideration when selecting the lens. The Computar M3Z1228C-MP zoom lens is used in this system. Get better images while meeting the imaging requirements of different sizes of glass bottles in the field of view.
At the same time, a dedicated signal processor is designed to receive the trigger signal from the sensor and forward the trigger signal to the camera. In addition, the reject information sent by the receiving image processor can be automatically calculated by the encoder to delay the number of rows and accurately output the reject trigger signal.
The overall hardware structure is shown in Figure 2: 
Figure 2 Hardware System Structure 
Figure 3 Detection algorithm flow chart 
Figure 5 system operation main interface
The use of more than one year of on-site use shows that the system can fully meet the requirements for the quality inspection of medicinal glass bottles, greatly improving its production efficiency, and at the same time providing an effective guarantee for the production of qualified products. However, in order to provide perfect detection equipment, we need to continue to research in the following aspects while ensuring the current detection speed: for the study of the threaded part of the bottle mouth; for the study of irregular scale partial defects; small for the bottom part of the bottle Defect research; Improve the statistical analysis of the wrong types of defective glass bottles and provide reference for the factory to improve the production process