dc.contributor.author |
Shuva, Fazleh Rakib |
|
dc.date.accessioned |
2021-09-15T06:19:51Z |
|
dc.date.available |
2021-09-15T06:19:51Z |
|
dc.date.issued |
2021 |
|
dc.identifier.uri |
http://dspace.daffodilvarsity.edu.bd:8080/handle/123456789/6145 |
|
dc.description.abstract |
The best way to automatically detect damaged images of accident vehicles. Image-based damage vehicles can effectively reduce the cost of an insurance claim. It will be benefited for the user and the insurance company. In this process user or vehicle, the owner takes a picture of damaged vehicles using mobile phone. then upload these pictures to the system. Automatically insurance claim process is done. However, this kind of solution is very difficult and challenging work. We have to work outdoor also. when an accident happened vehicle can be damaged .to identify the damaged part we have to submit this damage picture in our system and it will show us the damaged part of the vehicle.
This process is important nowadays. It has a huge scope for automation. In this paper, we learn about car damage classification. We use Deep Learning for this goal. We propose to use the CNN models of unveiled vehicles that are used to get the damaged picture.
If the pictures are damaged then the CNN model identifies the damaged part of the vehicle. And if the pictures are not damaged then the CNN model identifies no damaged The edges of the picture that are absent in the CNN model projection can be viewed as vehicle damage. Automatic photograph-based vehicle damage detection systems will provide the basis for the design. Moreover, we hope that our approach will provide the basis for interesting future research. |
en_US |
dc.language.iso |
en_US |
en_US |
dc.publisher |
Daffodil International University |
en_US |
dc.subject |
Image processing |
en_US |
dc.subject |
Image processing equipment industry |
en_US |
dc.subject |
Image processing--Digital techniques--Software |
en_US |
dc.title |
Car Damage Recognition Using Image Processing |
en_US |
dc.type |
Other |
en_US |