Indian Currency detector Android App
It is a app used to detect indian currency using flutter and tensorflowlite
The Flutter Application for the app See The Money which recognizes Indian Currency notes from 10 to 1000. The uploaded package is the full Eclipse Project that went into creating the app. The .apk file may be found in the bin folder which can then be directly ported to an Android phone.
The app has a custom touch-based interface for use by the visually impaired. Once the app is started (presumably using Screen Reader or some related application), the back camera of the phone automatically starts. The note should be held in the view of the camera (the visually impaired can get used to this after practicing a few times) and the phone should be held fairly steady for a few seconds, after which the photo will be taken. Then 3 situations may arise:
Case 1 (Easy Result) - The photo quality is good and the note is easily identified. The app reports the result after which the user can shake the phone 2-3 times and resume camera feed.
Case 2 (Reasonable Doubt) - Due to lower quality of the note/image or occlusion, the app detects multiple possibilities for the note. The message "Please take the photo again" will be played after which the user should take a different side of the SAME note and take another photo. This will have to be done a maximum of three times if needed.
Case 3 (No Chance) - Due to the note not being there in the image or an extremely low quality of the image, the app cannot identify any possibility for the note. The message "Bad Image! Please adjust the note" will be played.
To close the app, the user must touch the screen. Therefore it is recommended to hold the side of the phone while testing the app.
The app has a custom touch-based interface for use by the visually impaired. Once the app is started (presumably using Screen Reader or some related application), the back camera of the phone automatically starts. The note should be held in the view of the camera (the visually impaired can get used to this after practicing a few times) and the phone should be held fairly steady for a few seconds, after which the photo will be taken. Then 3 situations may arise:
Case 1 (Easy Result) - The photo quality is good and the note is easily identified. The app reports the result after which the user can shake the phone 2-3 times and resume camera feed.
Case 2 (Reasonable Doubt) - Due to lower quality of the note/image or occlusion, the app detects multiple possibilities for the note. The message "Please take the photo again" will be played after which the user should take a different side of the SAME note and take another photo. This will have to be done a maximum of three times if needed.
Case 3 (No Chance) - Due to the note not being there in the image or an extremely low quality of the image, the app cannot identify any possibility for the note. The message "Bad Image! Please adjust the note" will be played.
To close the app, the user must touch the screen. Therefore it is recommended to hold the side of the phone while testing the app.
