Artificial intelligence has been keeping up with the fast-paced world of agricultural innovations both conceptually and theoretically. In the more recent developments in AI, it has been noticed that artificial intelligence is presently dealing mostly with the discovery and not the distance mapping process concerning viable resources, and so, with a stronger functional perspective, the authors have made attempts towards AI-based systems that could potentially augment the meticulousness of the AI developmental aspect where fallow land needs dealing with. Given a vertical mapping of ground data with the help of satellite images, land use can be identified. Adding to this, a horizontal mapping using RGB frames stands to improve model performance to determine if a given area of land is feasible for agricultural development. The paper proposes a CNN model that can interpret and categorize agricultural areas based on GIS data. The primary application of this research is to facilitate farm land use cases to improve productivity and land viability.