The selection of an appropriate change detection method is considerably significant in producing a high-quality change detection remotely sensed product. Many change detection techniques have been developed in a trial to achieve accurate results by using different remotely sensed data, different study areas, and different classification algorithms as a base for many change detection analysis techniques. Artificial Neural Networks (ANNs) is considered one of the most promising advanced artificial intelligence techniques to be applied in the classification analysis. This research generates a new approach for enhancing change detection analysis. It develops an advanced supervised ANNs-based change detection to detect temporal changes in palm trees plantation in the study area for two periods depicted in different multi-source, multi-temporal and multi-spectral satellite dataset. Preceding the classification analysis, a Median Spatial Convolution Filtering (MSCF) was applied for each temporal image. Additionally, a Normalized Difference Vegetation Index (NDVI) was used to transform multispectral data into a single gray scale image band representing vegetation distribution in each anniversary date image. The NDVI band was masked and infused into the ANNs classification procedures. Furthermore, This study adopts two main change detection algorithms, the post-classification comparison and the change versus no-change binary mask. Two change detection statistical matrix reports were produced, as well as two binary difference (decline or growth) maps for the two periods of study. Three assessments of accuracy were computed for the proposed ANNs method, as well as for the resultant binary difference maps. All overall accuracies exceeded 97% with Kappa coefficient above 0.96, providing a promising method for change detection enhancement analyses.