Segmentation Base Advance Approach for Detect and Classify Leaf Diseases

1. INTRODUCTION

Image processing is a process of taking an image as an input, perform some operation it, in order to manipulated the image which include enhancing reducing rotating etc. or extracting some useful information from it and producing the desired output. The purpose of agriculture is not to feed ever growing population but it is an important sources of energy and a solution to solve the problem of global warming. Plant diseases are extremely significant, as that can adversely affect both quality and quantity of crops in agriculture production. Plant diseases diagnosis is very essential in earrlier stage in order to cure and control them. Most leaf diseases are caused by fungi, bacteria and viruses Fungi are identified primarily from their morphology, with emphasis placed on their reproductive structures. Bacteria are considered more primitive than fungi and generally have simpler life cycles. With few exceptions, bacteria exist as single cells and increase in numbers by dividing into two cells during a process called binary fission viruses are extremely tiny particles consisting of protein and genetic material with no associated protein. There are many types of leaf diseases. Rusts:-Rusts are plant diseases caused by pathogenic fungi of the order pucciniales. Rots:-Rot is a name used for various diseases of cultivated plants caused by fungi or bacteria, producing dark brown discoloration and decay in the leaves of fruit and vegetables: A diseases of the apple, pear and quinces caused by a fungus (Physalosapora cydoniae ) Blight:-Blight is a rapid and complete chlorosis, browning, and the death of plant tissues such as leaves, branches, twings, or floral organs. Fire blight of pome fruits, caused by the bacterium Erwinia amylovora (Burrill) winslow et al., is the most severe diseases of pear and also is found in apple and raspberry, among others. Leaf spot:-leaf spot is a common descriptive term applied toa number of diseases affecting the foliage of ornamentals and shade by fungi, but some are caused by bacteria. Some insects also causes damage that appears like a leaf spot diseases. Downy mildews :-Downy mildew is a micro-biotic fungal infection that occurs in cool moist weather and affectes many different plants. It is most often recognized by a yellow or faded spotting that appears on the leaves and is a particular problem in the northeast United States. Powdery mildew:-Powdery mildew is a common fungal diseases that affects many different plants. It is caused by a wide array of fungi of the Erysiphales order. Pososphaera xanthi is the most common reported cause. Powdery mildew is easy

Now a days, there are different types of detect leaf diseases are available. And we can classify them into several categories as follows:- • Support vector machine (SVM) • Gray level co –occurrence matrix (GLCM)

In this section we briefly explain the concepts, advantages, drawbacks of the detect leaf diseases. Support vector machines (SVMs) are supervised learning models with associated learning algorithm that analyze data used for classification and regression analysis. Support vector machine is based on maximizing the minimum distance from the separating hyper plane to the nearest sample. Only binary classification is supported in basic SVM, but for multiclass classification case extension can be possible[8]. In these extensions, additional constraints and parameters are added to optimization problems for handling the separation of the different classes. SVM is a binary classifier that means the class labels can only take two values ±1. To get M-class classifiers, set of binary classifiers are constructed in this way ……… and each is trained for separating one class from the rest. The function returns the signed real value that can be interpreted as distance from separation of hyper plane to point x.. The larger the value the more confident one is that the point x belong to the positive class. Hence, assign point x to the class whose confidence value is largest for this point.

• Its prediction accuracy is high. • Its working is robust when training examples contain errors. • Its simple geometric interpretation and a sparse solution. • Like neural networks the computational complexity of SVMs does not depend on the dimensionality of the input space.

• This classifier involves long training time. • In SVM it is difficult to understand the learned function (weights). • The large number of support vectors used from the training set to perform classification task.

The GLCM features are extracted from the image. Gray-Level Co-Occurrence Matrix (GLCM) is the statistical method of investigating texture which considers the spatial relationship of pixels [2]. The GLCM functions characterize the texture of images by computing the spatial relationship among the pixels in the images. The statistical measures are extracted from this matrix. In the creation of GLCMs, an array of offsets which describe pixel relationships of varying direction and distance have to be specified. In the proposed method, four features are extracted which include contrast, energy, homogeneity and correlation. Let Pij represents the (i,j)th entry in the normalized Gray-Level Co-Occurrence Matrix. N represents the number of distinct gray levels in the quantized image. The different features extracted are defined [4] The algorithm for K-Means clustering is: i. Initialize the center for each cluster. ii. Notify the nearest cluster to every data value. iii. Set the location of each cluster to the mean of all data points fitting in to that cluster iv. Repeat steps ii and iii until all data elements are converged to one of the cluster center. The performance of the clustering technique is measured using precision, recall and F1 values

First of all we load the input image from the leaf snap database

Step 2: Read and display an RGB image, and then convert it to grayscale.

Finally using parameters detect leaf diseases and output is here.

The terms true positives, true negatives, false positives, and false negatives compare the results of the classifier under test with trusted external judgments. The terms positive and negative refer to the classifier's prediction and the terms true and false refer to whether that prediction corresponds to the external judgment.

Recall :-

Accuracy=TP+TN/TP+TN+FP+FN

Step 1: Read image. Step 2: Applied pre processing step using median filter. Step 3: Identify applied the region of interest (ROI) meaningful content. Step 4: Erosion morphological operator to allow meaningful content. Step Step 5: Create a feature matrix. Step 6: Applied the feature matrix.

The paper presents a comparative study regarding leaf diseases in order to evaluate different types of image segmentation method and SVM the advantages and disadvantages for each of the approached algorithms. We have presented a new approach for leaf diseases or image segmentation (thresholding and morphological operators) SM classifier for allowing detect leaf and find diseases or solve diseases. It has been concluded that thresholding operator and SVM classifier is having a significant importance in digital image processing and also work on other parameters PRESION, RECALL & ACCURACY

Ramalakshmi, K. H. Mangaleswari, T. Kaviya (2017). “Detection of Leaf Diseases and Classification using Digital Image Processing” 2017 International Conference on Innovations in Information, Embedded and Communication Systems (ICIIECS) 2. Vijayashree T. & Dr. A. Gopal (2017). “Leaf Identification for the Extraction of Medicinal Qualities Using Image Processing Algorithm” International Conference on Intelligent Computing and Control (I2C2). 3. Dr. P. Tamilselvi & Dr. K. Ashok Kumar (2017). “Unsupervised Machine Learning for Clustering the Infected Leaves based on the Leaf-colours” 2017 Third International Conference on Science Technology Engineering & Management (ICONSTEM). 4. Anand R,Veni S,Aravinth J (2016). “An Application of image processing techniques for Detection of Diseases on Brinjal Leaves Using K-Means Clustering Method" Fifth International Conference On Recent Trends In Information Technology. 5. Asem Khmag, S.A.R. Al-Haddad, Noraziahtulhidayu Kamarudin (2017). “Recognition System for Leaf Images Based on Its Leaf Contour and Centroid” 2017 IEEE 15th Student Conference on Research and Development (SCOReD). 6. Chaitali G. Dhaware & Mrs. K. H. Wanjale (2017). “A Modern Approach for Plant Leaf Disease Classification which Depends on Leaf Image Processing” 2017 International Conference on Computer Communication and Informatics (ICCCI -2017), Jan. 05 – 07, 2017, Coimbatore, INDIA. 7. Dr. Sujatha S. & Preeti Kumari : “Recognition System for Leaf Images Based on Its Leaf Contour And Centroid” International Journal of Science, Engineering and Technology Research (IJSETR). 8. Neelu Arora & Mrs. G. Sarvani : “A Review Paper on Gabor Filter Algorithm & Its Applications” International Journal of Advanced Research in Electronics and Communication Engineering (IJARECE). 9. Monika, Renu Bala : “Image Edge Detection using Discrete Wavelet Transform” International Journal of Innovative Research in Computer and Communication Engineering. Manoj Manmode. 11. https://en.wikipedia.org/wiki/ Precision_and_recall 12. http://leafsnap.com/dataset/

PG Scholar, EC Department, KITRC, Kalol, Gujarat