OpenCV code for automatic micro egg detection in Android

Description

The algorithm pipeline is divided into two stages: filter detection and bead detection. Initial prototypes were designed in MATLAB and fully implemented in OpenCV for Windows and Android.

The algorithm was implemented and tested on Android with OpenCV. For the toy example given below, there were 33 connected components detected, with 20 true positives, 13 false positives, 4 false negatives, and (4608×3456)-33 true negatives (corresponding to all the remaining pixels). This yields an sensitivity of 83.33% and specificity of 100.00%.

Automatic detection of beer labels

MATLAB code for Automatic Coin and Bill Detection

https://opencv.org/