International Journal of Computer Applications |
Foundation of Computer Science (FCS), NY, USA |
Volume 181 - Number 2 |
Year of Publication: 2018 |
Authors: Khaled Marji Alresheedi |
![]() |
Khaled Marji Alresheedi . Fusion Approach for Dates Fruit Classification. International Journal of Computer Applications. 181, 2 ( Jul 2018), 17-20. DOI=10.5120/ijca2018917415
Depending on the type of the date fruit, dates have many features that should help in recognizing and classifying a date. The main features of the date fruit are the color, texture and the shape of date fruit. This article thus analyzes image based date fruit classification and recognition. In this regards, this paper has two contributions, firstly, as there is no standard dataset available, a dataset of 9 date’s classes is constructed. This dataset presents an interesting challenge for computer vision algorithms. Secondly, Gabor features and Color Layout features are used. These features are then fused to increase the classification performance. The proposed Color Layout and Gabor approach achieves an acceptable performance of more than 88% correct detections.