Object Recognition with UAV

In recent years, unmanned aerial vehicles(UAVs) have been used for numerous applications such as defence industry, urban planning, agriculture, emergency response and disaster management. Like most of the other widely used machines, increasing UAVs autonomy and independent decision making is literally a must in order to improve eciency, safety and accuracy of the devices. A UAVs autonomy directly a ected by its ability to model the outer world which is now more successful with the improvements of the computer vision. In the last few years, Convolutional Neural Networks(CNNs) have showed a remarkable performance in terms of object recognition. This paper details my senior project which is “Object Recogninition with UAV”. Mainly we can divide

this project into 2 part. First part is controlling the drone from computer and taking photographs and second part is processing that photographs, recognizing objects and showing them. For the rst part of the project Node.js is been used and for the second part of the project Tensor ow is been used.