All Categories
History
Price (excluding tax)
■Object recognition model suitable for FA DeepSky (object recognition) is an all-in-one and simple-to-operate AI image inspection software that can be easily operated by on-site users. Adopts a model suitable for use in FA. Recognize parts and defects with high accuracy. All-in-one: "Image data collection → learning → judgment" can be performed with a single PC. No need to make detailed settings adjustments: just circle the part you want to find in the captured image and click the "Learn" button. The AI automatically adjusts the setting parameters and starts recognizing the target object after 10 minutes (*), eliminating the need for the difficult setting work that used to be necessary. Unlike conventional procedural (rule-based) image processing, AI image processing is easy to set up, and is effective in making decisions under complex and easily changing conditions. *The required study time varies depending on the subject, and can range from 5 minutes to several hours. ■What can be detected There are various types of object recognition mechanisms. For example, there are those that have high recognition accuracy and require a high-performance PC, and those that do not have high recognition accuracy but can be processed at high speed. DeepSky is a type of AI with high recognition accuracy and adaptability tailored to FA applications. Therefore, a high-performance PC costing around 200,000 to 300,000 yen is required. The inspection time is about 0.2 to 0.5 seconds. ・Defect inspection: stains and tears on cloth products, textured molded products, mounting boards, etc. ・Inspection of scratches: dents, scratches, etc. on press-formed products ・Counting/Quantity Judgment: Managing the number of parts, checking for omissions and orientation of boxed products, etc. ・Assembly confirmation: Wiring color of wire harness, incorrect placement of parts, half-fitting, etc. ・Presence confirmation: Check attachments, included items, installed parts, etc. ・Classification of objects: confirmation of product combinations, etc. ・Status monitoring: confirmation of rotation/stopping of objects, determination of abnormal conditions, etc. ・Other uses other than FA: stamp and coin identification, topography, mask wearing, etc.
You can search for other models from each index. The displayed value is the value of the currently selected part number.
CSV (Excel) data storage
Count number
I/O input/output
Image storage
Number of images used for learning
RS232C communication or socket communication
Remarks
Study time
Types of detection targets that can be registered per product
Part Number
DeepSkyProduct
AI visual inspection DeepSkyHandling Company
Skylogic Co., Ltd.Categories
Image | Price (excluding tax) | CSV (Excel) data storage | Count number | I/O input/output | Image storage | Number of images used for learning | RS232C communication or socket communication | Remarks | Study time | Types of detection targets that can be registered per product |
---|---|---|---|---|---|---|---|---|---|---|
![]() |
Available upon quote | Divide and save by month (or day) by variety. | 0~1,000 pieces | Optional intelligent I/O allows for type switching, inspection trigger input, and OK/NG/BUSY output. | Categorize and save in OK and NG folders for each type. (JPG, PNG) | Approximately 10 to 1,000 images (The clearer the characteristics of the detection target, the fewer images and less time can be learned.) | Switch types, start inspection, and output pass/fail judgment. It also returns the label name and number of detected objects. | The type here refers to the classification name used when annotating, such as "screw," "desiccant," and "tomato." If you increase the number of detection target types beyond 100, the detection rate may decrease. | 5 minutes to several hours | 1~1,000 types |
Reviews shown here are reviews of companies.
Reviews shown here are reviews of companies.