Skip to content

Determination of defects through chromate process result values ​​and image data

Notifications You must be signed in to change notification settings

SiwooPark00/Quality_Anomaly_Detection

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

4 Commits
 
 
 
 

Repository files navigation

Quality_Anomaly_Detection

The goal of this project is Determination of defects through chromate process result values ​​and image data.
This is a process of applying a rust-preventive film using dichromate, which is used for high-end products that require gloss.
If the concentration of the solution is low, the result of the white chromate process does not appear, and if the concentration of the solution is high, the galvanized thickness is reduced by more than 1𝜇𝑚
Due to environmental regulations, the reaction product must be less than hexavalent chromium.
The dissolution and precipitation reactions involved in the process are not simple in their reaction structure and pathway, so it is important to create an appropriate reaction environment and conditions.

There are two types of data: image and numerical data.
image
image
I created each model based on both types of data.
As a result, The best model's accuracy(on test data) is 100.0%.
You can download data from the below link.
https://www.kamp-ai.kr/aidataDetail?AI_SEARCH=%ED%81%AC%EB%A1%9C%EB%A9%94%EC%9D%B4%ED%8A%B8&page=1&DATASET_SEQ=25&EQUIP_SEL=&GUBUN_SEL=&FILE_TYPE_SEL=&WDATE_SEL=

About

Determination of defects through chromate process result values ​​and image data

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published