Now, a plant's pic will tell you its disease!

These students can detect, quantify and identify a plant’s diseases and give possible solutions by just clicking its leaves.

Three final year students from electronic and telecommunications stream of Bharati Vidyapeeth’s College of Engineering for Women achieved this feat using digital image processing with the help of robotics and software.

The students Priyanka Padhye, Sanam Shikalgar and Kanval Rajanai, claim that though they have used hardware and software combination to detect the diseases, there is a scope to create an android mobile application  for this purpose. Farmers can use the the app and detect the diseases on their own by simply clicking the photos, they say.

“The robotic system is used for real-time crop monitoring and used for image acquisition. It is controlled via bluetooth and the image will be transferred to the host laptop,” said Padhye. She said that with the use of software they then identify mostly green-coloured pixels. Next, these pixels are masked based on specific threshold values. The other additional step is that the pixels with zero red, green and blue values and the pixels on the boundaries of the infected cluster (object) are completely removed. Further they train samples and using SVM classifier identify the diseases.

Another student Kanval Rajani said that on an experimental basis, they used mango tree and took the pictures of the leaves and identified that the it had Anthracnose (a fungal infection). She said that since they do not have subject knowledge of botany, they have taken the help of experts and helped them to upload the information of diseases and their solutions in the developed software. Anthracnose can cause a direct loss of fruit and if left untreated in harvested fruit, the blemishes it produces can make mangoes hard to market. In the invention of the particular system, the girl students were guided by professor ST Khot, who is the head of department of E&TC department.