Field Programmable Gate Arrays (FPGAs) are semiconductor devices that are based around a matrix of configurable logic blocks (CLBs) connected via programmable interconnects. FPGAs can be reprogrammed to desired application or functionality requirements after manufacturing. This feature distinguishes FPGAs from Application Specific Integrated Circuits (ASIC), which are custom manufactured for specific design tasks. The purpose of this paper is to study and compare various FPGA neural networks and determine the most effective among them in terms of energy efficiency. Even though various research studies worked on different FPGA neural model designs, very few researchers have considered comparing these designs to determine energy efficiency. This paper provides Ad-MobileNet as an efficient solution, a system constructed on the conventional MobileNet topology that utilizes fewer computer complexity while achieving classification techniques superior to some well-known FPGA-based platforms․ Because of this, we've investigate the Ad-MobileNet model with larger datasets and a range of uses as part of continuing research.