Plant Disease Detection Using Deep Learning

Abstract​

Crop diseases are one of the major issues, but their identification is difficult due to the lack of required infrastructure. Plant diseases affect farmers whose livelihood depends on the crops and also it increases the vulnerability of food security at the large scale. Plant disease identification is very important because it affects the growth of the plant species. Usage of pesticides reduces the ability to fight back. The plant diseases are detected using Deep Convolutional Neural Network trained and added into the database of leaves from different plants.The dataset consists of 38 disease classes and one background class stanford open dataset. The proposed model predicts from the image of a leaf if it is diseased or not and also gives the name of the disease predicted.

In Deep learning, CNN is an algorithm that takes an image as an input and it assigns importance to the objects in the image and we can distinguish one image from the other.

Architecture used is resnet architecture. We have many types of resnet architecture i.e. concept is same but with different number of layers, for example we have ResNet-34, ResNet-50, ResNet-18, ResNet-101 etc. Here, In this paper we are using ResNet-50 architecture, since it is a variant of Resnet model that has 48 convolution layers along with one average pool and one max pool layer. The results are deployed into the cloud(AWS) where the data can be fetched whenever required. Accuracy of the system is around 97-98%.

INTRODUCTION

Indian economy is highly dependent on agriculture therefore detection of plant diseases becomes very important.. In this paper we use image classification concepts to recognise different plant diseases.Plant And also we have to halt the waste of financial and many other resources.

Solution to this problem is as follows:

We have built a model that uses deep learning to detect and classify by taking an image of a leaf and classifies leaves using CNN predict the type of disease that might have affected the plant.

The process proceeds as follows:

1. The leaf is detected in a given image.
2. Extracted leaf is run through a CNN classifier to identify which the leaf belongs to which plant.
3. The leaf image is then used to predict for which disease class it belongs to.

The Deep Learning (DL) approach is a subset of Machine Learning (ML) ,has networks capable of learning unlabelled data(Unsupervised Learning). This includes several developments like hand written text recognition, Backpropagation, chain rule etc.However, in the next phase, architectures were developed and used for many applications like healthcare sector, finance and earthquake predictions.Among those architectures,ResNet-50 architecture is used.ResNet-50 is a deep residual network and are mainly used for image classification.It uses the Skip connection from an earlier layer to a next layer to add the output.When these architectures started to evolve, The research has been done and applied to image recognition and classification.Some of these architectures have been or various agricultural applications.For example classification of leaves was carried out by using CNN classifier CNN has input layer, convolution layer, pooling layer and classification layer and output layer. when compared to other algorithms diseases should be diagnosed very accurately.

Save your time!
We can take care of your essay

• Proper editing and formatting
• Free revision, title page, and bibliography
• Flexible prices and money-back guarantee

Place Order

LITERATURE REVIEW

[1] Deep Neural Networks are used to detect and classify plant leaf diseases: In this paper, Deep learning technique is used to detect the plant diseases and classify them by examining leaf of a given plant.

[2]Recognition of Plant leaf Using a Convolution Neural Network: In this paper we use YOLOv3 object detector, which is used to extract leaf from a given image and it is been analysed through ResNet-18 models.

[3] Using Deep learning methods to classify Banana leaf diseases: In this paper, Deep learning techniques are used to classify the banana leaves diseases. The architecture used here in this paper is LeNet architecture that is used as a convolutional network architecture in order to classify the images.

[4] Image Recognition using Deep Residual technique: In this paper, here we have a residual learning framework which is deeper than the framework used before. These frameworks formulate the layers again as learning residual functions with the reference to inputs of the layers, instead of learning other unexploited functions.

[5] Deep Neural Networks are used for large scale plant classification: In this paper, we discuss about the possibility of deep learning techniques to apply for classification of plants and it is used for large scale biodiversity monitoring and also the plant classification using convolutional network architectures like ResNet50 gives higher accuracy compared to other plant classification applications​.

[6] Detection of Plant leaf diseases using segmentation of and soft computing techniques: In this paper,an algorithm is used to segment the images i.e, segmentation technique to classify and detect the leaf diseases.

[7] ​Leaf Recognition Algorithm to classify the Plant Using Probabilistic Neural Network: In this paper, To implement leaf recognition algorithm, we use PNN. 11 leaf characters are obtained. PNN is trained to classify 32 kinds of plants for 1800 leaves which gives accuracy greater than 90%.

CONCLUSION

Using deep learning techniques, we executed a method to detect and classify the plant diseases from the leaf images. The obtained representational model can differentiate between healthy leaves and unhealthy leaves containing 38 different diseases, that can be visibly acknowledged. The obtained results after the experiment have achieved accuracy precision between 91% and 98%, for separate class tests. The obtained accuracy for this trained model is 97.5%.

REFERENCES

1. Aravindhan Venkataramanan,Deepak Kumar P Honakeri,Pooja Agarwal .”Plant Disease Detection and Classification Using Deep Neural Networks ” International Journal on Computer Science and Engineering (IJCSE).
2. Jeon, Wang-Su, and Sang-Yong Rhee. “ Using CNN to detect and classify the Plant leaf diseases.' International Journal of Fuzzy Logic and Intelligent Systems 17.1 (2017): 26-34.
3. Amara, Jihen, Bassem Bouaziz, and Alsayed Algergawy. '​Using Deep learning method to classify Banana leaf diseases​.' BTW 2017.
4. He, Kaiming, et al. 'Deep residual learning approach for the recognition of images.' Proceedings of the IEEE conference on computer vision and pattern recognition. 2016.
5. Heredia, Ignacio. '​Deep Neural Networks are used for large scale plant classification​' Proceedings of the Computing Frontiers Conference. ACM, 2017.
6. M. B. Riley, M. R. Williamson, and O. Maloy, “Plant disease diagnosis. The Plant Health
7. vijai singh,A K Misra,”Detection of plant leaf diseases using image segmentation and soft computing techniques”,Information processing in agriculture,vol.4 issue1 March 2017.
8. Stephen Gang Wu,Forrest Sheng Bao,”​A Leaf Recognition Algorithm for Plant Classification Using Probabilistic Neural Network​”,IEEE International symposium on signal processing and information technology,2007
9. Munisami, Trishen, et al. ' We use shape features and colour histogram with KNN classifiers to recognize the Plant leaf.' Procedia Computer Science 58 (2015): 740-747.
10. Al-Hiary, Heba, et al. ' Detection and classification of plant leaf diseases will be accurate and fast.' International Journal of Computer Applications 17.1 (2011): 31-38.
11. Sladojevic, Srdjan, et al. ' Recognition of plant diseases are based on DNN for leaf image classification.' Computational intelligence and neuroscience 2016 (2016).
12. Lee, Sue Han, et al. 'Plant Identification System based on a Convolutional Neural Network for the LifeClef 2016 Plant Classification Task.' CLEF (Working Notes). 2016.]Redmon, Joseph, and Ali Farhadi. 'Yolov3: An incremental improvement.' arXiv preprint arXiv:1804.02767 (2018).
13. S. A. Miller, F. D. Beed, and C. L. Harmon, “Plant disease diagnostic capabilities and networks,” Annual Review of Phytopathology, vol. 47, pp. 15–38, 2009.
14. J. G. Arnal Barbedo, “Digital image processing techniques i.e, digital learning for detecting, classifying the plant diseases”, SpringerPlus, vol. 2, article 660, pp. 1–12, 2013. Instructor,” 2002.