Satellites are transforming how we monitor our environment. They provide a vast amount of data that can be used for many purposes. This study introduces a novel work that uses deep learning to analyse satellite images and identify fish farms, shrimp farms, and saltpan ponds. The freely available high resolution Google earth images dataset is used for classifying the pond images. CNN based deep learning models Yolo v7 and YoloV8 explored to train on 20000 labelled images distributed among four classes, we achieved high accuracy in distinguishing between important aquaculture farming practices features. Testing with a large dataset shows that this system is effective and can be used to monitor large areas quickly and reliably