スカイロジック Product Lineup

It has been reported that there was an incident where products with similar package types were shipped with incorrect labels. In many cases where multiple types are manufactured in a flow production system, visual inspection systems play a crucial role. 【Inspection Settings and Results】 By using the "Comparison with Master Image" feature of EasyInspector, judgments could be made in less than one second. The color judgment tolerance was set to 69. As shown in the image above on the right, differences from the master image are detected in red. 【Software Used】 Software Used: EasyInspector310 Number of Inspection Points: 1 The current 'EasyInspector2' color package can be inspected using the "Comparison with Master Image" feature.

It seems that the inspection involving a mix of letters and numbers was difficult to determine at the time. With improvements in lighting and settings, we were able to read with higher accuracy. This request was made in 2015, but the OCR reading function in the current EasyInspector has been enhanced, increasing its accuracy, allowing us to retain images and record content in a CSV file. 【Inspection Settings and Results】 By using EasyInspector's "Character Recognition (OCR)" function, we detected readings from 16 printed locations, achieving a probability accuracy of about 90%, and we were able to make determinations in under 3.41 seconds. A 3-megapixel camera provided the most stable inspections. It is indeed easier to misdetect when letters and numbers are mixed; for example, "8" and "B" are characters that are prone to misdetection. However, when "8" is among the numbers and "B" is among the letters, we were able to read them relatively accurately and stably. Although the number of inspection frames increases, we believe that separating letters and numbers and setting inspection frames accordingly will enable stable inspections.

The AI general-purpose appearance inspection software "EasyInspector2" has newly added the "AI Segmentation" feature. The "AI Segmentation" feature performs segmentation (region division and labeling of each pixel in the image) using deep learning within the image. It calculates the ratio of pixels detected by segmentation within a specified polygonal area and makes pass/fail judgments. This feature can mainly be used for the following applications: - Tank water level - Area of metal processing surfaces and coating area - Measurement of crop growth (size) - Detection of lesions in photos and measurement of their area - Checking for coating gaps in sealants and adhesives - Monitoring the amount of smoke emitted By using the AI Segmentation feature for judgment, it achieves: - Improved work efficiency: Significant time savings compared to visual inspection - Enhanced accuracy in quality control: Reducing human errors and stabilizing quality - Cost reduction: Lower labor costs and reduced costs due to suppression of defective products.

In conventional image processing, it is possible to inspect the degree of terminal insertion, but there are challenges such as limited connector shapes and the need for positioning. Therefore, we decided to use AI for the inspection. By training the AI to recognize uninserted, partially inserted, and fully inserted states, it becomes possible to identify them. While the AI object recognition function is sufficient for "detection," the real issue is how to make the judgment. AI object recognition establishes inspection by having humans create settings that search for and recognize defects, such as scratches or dents, from the overall workpiece—if even one defect is found, it is deemed a failure (NG). In other words, object recognition alone can identify an object, but it cannot make a judgment that it is acceptable just because it appears in a certain alignment. This is where the OCR function comes into play. *For more details, please refer to the related link (blog).