Renewable energy contributes significantly to power generation to tackle the energy demand. Renewable energy is obtained from solar, wind, hydroelectric, etc. Among these, solar energy is considered the best suitable energy in terms of cleanliness and directly converts sunlight into electrical power by solar photovoltaic (PV) module. Solar panels' randomly changing power output due to irradiance is the biggest problem with solar panels. The concept of maximum power point tracking (MPPT) techniques is introduced to tackle this non-linear behavior of PV and optimize the PV system's efficiency. Various MPPT techniques have been proposed based on conventional and intelligent methods. In this work, a novel image processing-based MPPT technique is introduced to increase the efficiency of PV. The irradiance level is accurately classified using the self-learned EfficientNetB0 deep learning model. The parameters of the EfficientNetB0 model are adjusted using Tuna Swarm Optimization. Results show that the tracking efficiency is higher than other intelligent MPPT techniques. Also, the classification accuracy of the proposed learning model is superior to conventional models.