Brightness preserving histogram equalization (BPHE) technique was used to enhance the features to discriminate three dates varieties (Khalas, Fard and Madina). Mean, entropy and kurtosis features were computed from the enhanced images and used in an Artificial Neural Network classifier. The classification efficiency of 4 sets of hidden neurons (5, 10, 20, and 30) was tested and the network with 5 neurons yielded the highest classification accuracy of 95.2%.
A, MANICKAVASAGAN.; THOMAS, GABRIEL; AL-YAHYAI, R.; and M, HEMA
"BRIGHTNESS PRESERVING HISTOGRAM EQUALIZATION (BPHE) TECHNIQUE FOR CLASSIFICATION OF DATES,"
International Journal of Image Processing and Vision Science: Vol. 1:
3, Article 11.
Available at: https://www.interscience.in/ijipvs/vol1/iss3/11