Syrup Bottle Full Checking Visual Inspection Machine For Pharma Packages

Syrup Bottle Full Checking Visual Inspection Machine For Pharma Packages

Inspection standards

Product background

The medicine bottle packaging appearance defect detection system is mainly used for fast and reliable detection of oral liquid glass medicine bottles, plastic bottles and plastic containers. Detection, etc. Guochen Robot focuses on the detection technology of the pharmaceutical industry, citing machine vision detection to improve the detection efficiency and accuracy of drugs, and reduce labor costs for enterprises. The machine vision defect detection of medicine bottles is mainly used in the production, packaging, sealing/sealing, labeling, coding, packing, etc. of medicines in the pharmaceutical process.

The traditional artificial visual defect detection method is time-consuming and labor-intensive. Based on machine time and deep convolutional neural network, the automatic detection of plastic drug surface defects is realized, which avoids the time-consuming and labor-intensive artificial visual inspection, reduces labor costs and greatly improves the inspection quality. The production efficiency is improved and the false detection rate is reduced, which is of great significance for widespread promotion.

Inspection details

KEYE TECH uses a 3-megapixel industrial camera to collect and extract image samples of special plastic bottles and preprocess them. Because plastic bottles usually have color-like characteristics, the HSV color space conversion is used to extract color features on the sample image, and the Otsu threshold is used to segment the feature part, which is beneficial to neural network training, reducing the difficulty of network training and improving the speed.

Product advantages

• Simple operation: fast modeling, wizard setting, intuitive user interface
• High detection accuracy: different accuracy levels can be set for different regions
• Low false positive rate: low detection false positive rate
• Detection speed: the fastest speed is 20000pcs/hour (different products are detected at different speeds)
• Bad archiving: The detected defects and bad pictures are archived in the designated folder, which can be traced back to the bad by the operator.

Application scenarios

The medicine bottle packaging appearance defect detection system can be applied to the on-line inspection of the appearance defects of oral liquid glass bottles, plastic bottles, plastic containers, beverage bottles, etc.

1. Support a variety of defect detection 3C product defect detection features to build an exclusive underlying convolutional neural network architecture. After uploading pictures of different defect data, workers will label all the defects that need to be detected in turn, and click training to complete the labeling. The characteristics of different defects can be accurately identified according to these characteristics. With the continuous increase of data, the accuracy rate also increases.

2. The part to be detected can be separated from the background without external interference, and the defect only needs to be larger than one pixel. The detection effect will not be interfered by many factors such as color and environment. The optimized algorithm can effectively solve the problems of reflection and high brightness of the product itself, and is perfectly suitable for identification and detection in complex backgrounds. The detection speed can reach the millisecond level, which greatly improves the work efficiency of the production line.

3. Accurately identify the qualified product pictures uploaded by users, and mark the 3C product or its surface text in the picture, so as to accurately locate the location information and integrity of the product, and identify the text content. It truly realizes non-contact and non-destructive automatic identification, and the identification speed can reach the millisecond level.

4. Intelligent management Users can manage the system platform, request multiple units at the same time, and the system will record the defects

5. High inspection accuracy You can flexibly combine different types of machines and solutions according to inspection requirements, workpiece size and inspection accuracy requirements.

6. Statistical analysis Real-time statistical test data to accurately record the current production status.

About Us

Development History

The technology center is composed of algorithm department, software department, optics department, hardware department and mechanical department. The algorithm team currently has 7 PhDs, ensuring that the AI algorithm has the ability to continuously update.

Organization Structure