In recent years, the protection of crops from the wild animals are very challenging factor. such as elephants, wild boars, moles, monkeys and many others which may often damage the crops by self-feeding or simply by running over the fields. Also, there are instances where human lives are lost due to the wild animal’s attack. The existing methods like electric fences, creating unpleasant noise and detecting the animals using RFID and LF tags which are injected into the animal ‘s skin is proving to be less efficient as few methods compromise on the safety of wild life and few on the human lives. Using image processing techniques, the animal is detected precisely and broadcasting the same information to forest officer and villagers. Also, the consequent actions like sedating the wild animal would need human intervention. The information would be passed on the nearest forest officers, thereby reducing the crop damages and minimizing the probable loss to the farmers.
In India, nearly 65% of the people are directly or indirectly dependant on agricultural sector for economic survival. The annual income of farmers is significantly influenced by the yield of the crops, which is continuously decreasing due to natural phenomena and poor technological advancement. The crops are damaged by birds and animals in the field. Many of the used methods result in extinction of the rare species. Therefore, there is a need to develop alternative techniques, such that it does not harm birds and animals physically as well as protects the crops. This project protects from birds and animals to reduce the loss of crops and thus helps farmers to reduce the risk of crop damage. In this project we have used Deep Convolutional technology to detect the animals and birds in the field. When the animals and birds are detected by camera and deep convolution process are done and the flash light and sound are given in the field to repellet the animals.
HARDWARE IMPLEMENTATION
SOFTWARE REQUIREMENTS
There is no Reviews about this Product
All fields are mandatory
