Geospatial images deliver a visual medium to recognize the prompt changes in the environment influenced due to seasonal changes. These changes extend their impact on the soil vegetation and crop growth patterns leading to an increase/decrease in agricultural production. Identification of seasonal changes and their impact using collected geospatial images within a specific time frame to predict land utilized for agriculture. Although identifying land cover utilization through spectral indices and bands are available, they generalize based on one particular region and may misclassify and overlap each other. The utilization of template matching enables in identifying the crop field area based on templates extracted and labelled. Masked images are acquired randomly from images or existing datasets labelled as fields cropped and barren. The specific region of interest is tiled and demarcated through Semi-supervised classification identifying the annual land crop cover. The extracted images are utilized as masked templates to identify the area of crop cover with an accuracy of 98%. The correlations of land cover helps in identify the specific Region of Interest (ROI) providing maximum and minimum utilization of cultivable lands.

地理空间图像提供了一种视觉媒介来识别受季节变化影响的环境的即时变化。这些变化扩大了对土壤植被和作物生长模式的影响，导致农业生产的增加/减少。使用特定时间范围内收集的地理空间图像识别季节性变化及其影响，以预测农业用地。尽管可以通过光谱指数和波段来识别土地覆盖利用，但它们基于一个特定区域进行概括，并且可能会错误分类并相互重叠。模板匹配的利用能够基于提取和标记的模板来识别作物田地区域。蒙版图像是从标记为已裁剪和贫瘠的字段的图像或现有数据集中随机获取的。通过识别一年生土地作物覆盖的半监督分类，对特定的感兴趣区域进行平铺和划分。提取的图像用作蒙版模板，以 98% 的准确率识别作物覆盖区域。土地覆盖的相关性有助于确定特定的感兴趣区域 (ROI)，从而最大限度地和最低限度地利用可耕地。