COMPREHENSIVE SOLUTION FOR INTELLIGENT MANUFACTURING OF MASKS

Solution 1

Layout 1 of Automatic Mask Production Line Scheme on line

Solution 2

Layout II of Automatic Mask Production Line Scheme offline

Advantage

Diverse detection content

Mask bare film inspection content:

Check the size of the mask (with an accuracy of ± 2mm), whether there are ear straps, missing welds on the ear straps, tying of the ear straps, residue of adhesive tape, presence of nasal bridge strips, dirt and foreign objects on the surface of the mask (which can detect hair strands); Mask flat bag inspection content:

Edge pressing, poor sealing, empty paper bags, residual tape, corners, multiple pieces, dirt and foreign objects, poor printing, etc

Efficient

The conventional production efficiency is 200pcs/min, Up to 500pcs/min

Powerful hardware combination

Using 1 or 2 sets of line scanning high-resolution industrial cameras and lenses to capture images of different defects; Multiple combinations of customized line scanning infrared light sources and line scanning white light sources are used to supplement the light, eliminating interference caused by the pattern of the mask itself and the printing of the mask packaging; Adopting AI edge servers to enhance computing power, thus ensuring the high-speed and smooth operation of AI algorithms and traditional visual detection algorithms; Adopting an I/O control unit based on FPGA core to ensure stable removal during high-speed operation

High detection accuracy

Unlike most manufacturers on the market who only use traditional visual detection algorithms, this product uses a combination of AI deep learning detection algorithm and traditional visual algorithms for detection: due to the varying grayscale values and various pattern changes on the surface and packaging of masks, and the grayscale changes at the junction position, it will have a significant impact on the detection of traditional algorithms, The diversity/inconsistency of defects is also difficult for traditional algorithms to achieve comprehensive detection, so AI deep learning combined with traditional algorithms is used to ensure the accuracy of detection

Quick landing for multiple mask detection

After long-term accumulation, this product has trained AI models for various types of bare masks and mask packaging on the market, forming various detection AI models that can achieve rapid landing of new products

Strong compatibility

The same detection device can detect multiple types of bare masks or mask packaging, and can achieve quick switching. It is compatible with adult masks, children's masks, and various customized masks, paper packaging, and paper plastic packaging

Low rate of missed and false detections

According to the classification of mask defects, a maximum missed detection rate of 0.3% and a false detection rate of 1% can be achieved for major defects

Serialized solutions

Our company also provides N95 mask naked film detection device and N95 vertical package detection device

Display of detection content

MES workstation