Ayurvedic Plant Identification using Image Processing and Artificial Intelligence

Plants play an important role in Earth’s ecology by providing sustenance, shelter and maintaining a healthy atmosphere. Some of these plants have important medicinal properties. Automatic recognition of plant leaf is a challenging problem in the area of computer vision. An efficient Ayurvedic plant leaf recognition system will beneficial to many sectors of society which include medicinal field, botanic research etc. With the help of image processing and artificial intelligence, we can easily recognize the leaf images.


I. INTRODUCTION
Ayurveda medicinal system is a very huge and better medicinal technique, apart from western medicine methods. Since the Vedic times Ayurveda is being practiced in India. Ayurveda is one of the oldest systems of Medicinal science that is even used today [3]. Ayurvedic plant identification has become a challenging process and an active area of research, it also plays a vital role in preserving physical and mental health of human beings [5]. As oxygen are the source of plants, they release it by the process of photosynthesis. Apart from this, the plants are used in variety of industrial applications such as herbs and ingredients in ayurvedic medicines [2]. People have been using plant as a traditional medicine and identification of the correct ayurvedic plants that goes into the preparation of a medicine is very important in ayurvedic medicinal industry. Cardiac disorders, Respiratory diseases, Fertility issues, etc can be cured using ayurvedic medicine [7]. So a precise identification of ayurvedic plant is crucial for proper treatment but there is a huge problem that most people cannot recognize these ayurvedic plants and thus are not able to take advantage of herbal power to cure disease [8]. Every plant on earth has some medicinal value, according to Ayurveda. Medicinal plant are those plants that are used in treating and preventing specific aliments and diseases that affect human beings. It is considered a form of alternative to allopathic medicine in the world. One of the major advantages of ayurvedic plant is that it does not have any side effects [2]

Review of Pre-Processing
The objective of the pre-processing step is to standardize the scale and orientation of the image before feature computation. The raw image is typically a color image oriented at a random angle and having a random size. The image is first converted to binary and gray scale forms. To make features rotation-invariant, the angle of the major axis of the leaf is extracted from the image and used to rotate it so that the major axis is aligned with the horizontal line [5].
• Gray scale conversion: The image converting to gray scale. The gray scaled images were subjected to the process of image contrast and intensity enhancement techniques and then stacked together as slices for further processing.

A. Shape Features
• Length of leaf: This is the distance between the two ends of the main vein of the leaf [1]. • Breadth of leaf: This is the distance from the left most point in a leaf to the right most point in the leaf. • Aspect Ratio: The aspect ratio of a leaf is the ratio of the length to its breadth. • Diameter: The diameter of a leaf is the maximum distance between any two points which lie inside the area covered by the leaf. Where G is the number of gray levels, P(i,j) is the probability distribution in the GLCM [5].

Color Features
Arithmetic Mean: Standard Deviation:

Skewness(Ѳ):
Kurtosis: Where M and N are the dimensions of the image, P(i,j) are values of the color on column i and row j [7].

V. Review of Classifiers
A. K-Nearest Neighbor(KNN) k-nearest neighbor algorithm is a method for classifying objects based on closest training examples in the feature space. k-nearest neighbor algorithm is among the simplest of all machine learning algorithms. Training process for this algorithm only consists of storing feature vectors and labels of the training images. In the classification process, the unlabelled query point is simply assigned to the label of its k nearest neighbors. Typically the object is classified based on the labels of its k nearest neighbors by majority vote. If k=1, the object is simply classified as the class of the object nearest to it. When there are only two classes, k must be a odd integer.
However, there can still be ties when k is an odd integer when performing multiclass classification. After we convert each image to a vector of fixedlength with real numbers, we used the most common distance function for KNN which is Euclidean distance [4].

B. Probabilistic Neural Network (PNN)
A probabilistic neural network (PNN) is a feed forward neural network . PNN is often used in classification problems. When an input is present, the first layer computes the distance from the input vector to the training input vectors. This produces a vector where its elements indicate how close the input is to the training input. The second layer sums the contribution for each class of inputs and produces its net output as a vector of probabilities. Finally, a compete transfer function on the output of the second layer picks the maximum of these probabilities, and produces a 1 (positive identification) for that class and a 0 (negative identification) for non-targeted classes. In a PNN, the operations are organized into a multilayered feed forward network with four layers. Four layers are Input layer, Hidden layer, Pattern layer/Summation layer and Output layer [1].

D. Decision Tree Classifier
Decision tree is an approach for predictive modeling used in data mining and machine learning, by which the data in data sets can be classified into classes. By using decision trees, the system eliminates half of the cases at each step. This is extremely beneficial because in a system like this, where the dataset is huge in number, the processing takes a chink of the total time. Hence, it has to be minimized wherever possible. The decision tree classifiers organized a series of test questions in a tree structure. The final result is a tree with decision nodes and leaf nodes. Each internal node represents a "test" on an attribute. Each branch represents the outcome of the test. Follow the appropriate branch based on the outcome of the test. It then leads us either to another internal node or to a leaf node [5].

E. Artificial Neural Network (ANN)
Artificial Neural Network have proven themselves as proficient classifiers and are particularly well suited for addressing non-linear problems like leaf classification. ANN is an interconnected group of nodes. Neural networks consist of multiple layers and the signal path traverses from front to back. Back propagation is where the forward stimulation is used to reset weights on the "front" neural units and this is sometimes done in combination with training where the correct result is known [7].