Overview / Usage

In modern Agriculture and Forestry purposes Plant Pathological Research is very much essential for the crop improvement. In every year, huge crop is damaged by pathogen attack. In most of the cases, Farmers and lay men suffers huge loss due to severe pathogenic infestation. For crop improvement, the first and foremost essential key point is diseases identification. After proper identification of diseases we will be able to treat the diseases. Most of the cases lab oriented research is time taking and very much expendable but instant identification of plant diseases with its causal organism by using modern technology is most suitable and cost effective and also authentic. This type identifying methodology will help us to identify the diseases with its proper control measures. Above 95 % accuracy and authentic identification is possible by using this methodology.

In my present research, I have taken two of the most economically important and valuable species such as Tea(Camellia Sinensis) and various species of Citrus as experimental basis. I have done the Canker and Blister Blight Diseases of the above mentioned plant species. And it shows almost 99% accurate identification of the disease in all the infected plants. It is very easy and simple technique which any farmer or lay men can use this methodology to Identify the symptoms and signs of the diseases. By using this methodology we can easily identify the various types of diseases which mostly occurs in Leaf, Stem, Root , Fruits and Tubers.

The plant species which we have taken for the demo of our ongoing research i.e. Tea and Citrus plants such as (Oranges) are economically very much valuable in the market and are always high in demand but cultivators or farmers are always facing a severe problems and huge monetary loss due to pathogenic attack. This also creates scarcity in the market which leads to the high prices of the Fruits and Tea.

This methodology finds it's application in Agriculture, Horticulture, Floriculture, Forestry and every aspect of crop improvement.

Methodology / Approach

Methodology and Architecture:-

1. Architecture: The model is having two variants, One built in Faster RCNN and the other in SSD Mobilenet (ssd_mobilenet_v2_coco). Final one is on the SSD Mobilenet, as SSD Mobilenet model is well supported by both OpenVino and TensorFlow Lite.
2. Back-end Framework: Intel Optimized TensorFlow.
3. OpenCV for the Computer Vision Algorithm building.
4. Dataset: High Quality Images taken from Google Images using Web Crawling. Actually this is one of the bottleneck's I am facing. Since going into the fields and clicking images is one of the most tedious and time consuming approach that's why I went for the images available on the internet.
   But, for my present approach I need high quality and a huge amount of data, as the classes are increasing. I will gather the images and create my own Dataset. In future that can be open sourced so that people can use that.
5. Log Loss: I am able to drop down the log loss up-to 0.0173 at a step count of 32,517 after 14 hours of training. In terms of accuracy, as the inference shows that it ranges between 85-95%, depending upon the image or image that has been provided for the inference.
   6: Mobile Application: The mobile application is being built for the Android Platform as of now. I am using TensorFlow Lite and Android Studio for building it.
6. Model Optimization and Inference on PC/Laptop or any other edge device other than Smart phone is being carried out by Intel Distribution of the OpenVino ToolKit.

Approach :-

1. Write a web crawler that will gather plant data from google images.
2. Label our dataset and create .xml file corresponding to each image.
3. Create train and test data directories. Create a script for generating the train.csv and test.csv for the data.
4. Generate the train.record and test.record files.
5. We will then train our classifier algorithm with the data using Tensorflow or Caffe using the Open Vino toolkit and create a model out of it.
6. Optimize our model to create an *.xml and *.bin file
7. Then we will create a setup using the Inference API so that it is easily gets optimized results on the CPU using the camera and finally we identify the plant disease given input as an image, a video or even a live camera feed.
8. Once the identification is done, then the model will provide other details like why the disease has happened, the possible cures, preventive measures etc.

Note: All the training is entirely done using Intel CPU. No GPU support whatsoever is used while training or Inference of the model. The system is built and trained entirely on Ubuntu 16.04. Intel Optimized Python is the building block of the system.

Technologies Used

Hardware Stack :-

1. Intel AI DevKit.
2. Intel RealSense Camera.
3. Intel AI DevCloud.

Software Stack :-

1. Intel Optimised Python.
2. Powered by TensorFlow 2.0
3. TensorFlow Lite.
4. Intel's OpenVino ToolKit for Computer Vision.

Other links

• Publication on Economics Time
• Earth Day Special Coverage
• Publication by GreenBiz as a Story
• Earth Day Tweet by Intel News
• Publication on FutureFive, NewZealand
• Publication on LiveMint
• Earth Day Publication
• World Agriculture Day

Collaborators