International Journal on Recent and Innovation Trends in Computing and Communication ISSN: 2321-8169 Volume: 6 Issue: 2 19 – 26 _______________________________________________________________________________________________ 19 IJRITCC | February 2018, Available @ http://www.ijritcc.org _______________________________________________________________________________________ Removal of Weeds in Agriculture field using Wavelet Transformation in Image Processing Dr.A.Senthil Rajan 1 Director, Computer Centre, Alagappa University, Karaikudi, India 1 Abstract: Weeds is a widely procedure to be remove in agriculture field. Weed species decreases the growth of the crop and reduce farm yields. Weeds grow along with main crops and compete with crop for sunlight, space and nutrients. To control weed species, a large number of pesticides and chemicals are used in agricultural fields, which results in drinking water contaminated and environmental pollution. Currently, therefore it is important to successfully identify the weeds from the crop to selectively spray herbicides to reduce wastage use of chemical. Wavelet is very popular tool in image processing algorithm. In this paper a new algorithm is developed for crop detection and management of weed. It is needed to measure the roots and size of the plant in terms of length for detection of crop. For this purpose segmentation and filtering techniques are used in a noise captured image. Keywords: Filters, Segmentation, Noise, Weeds, Agriculture __________________________________________________*****_________________________________________________ I. Introduction Segmentation of weed plant is one of the essential tool agriculture sectors. It is a process of unwanted plants in and around the agriculture field. The weed will affect the trees and plants growth. The weed also takes in all the nutrients in the soil and water content. To eradicate the weed is the challenging task for farmers. It needs lot of man power to remove the weeds. Today scenario it is difficult to adopt the labour especially of farm lands.Detecting the harmful weeds is the challenging task for the farmers. It is the process of partitioning the unwanted plants otherwise called weeds based on the separation of weed result, surface of plants and roots can be extracted, modeled, manipulated, measured and visualized. Edge detection is an essential task in computer vision. It covers wide range of application from segmentation to pattern matching. The edge measure parameters related to the plant stem and the roots are the first step before registration conventionally edge is detected according to same early brought forward algorithm like sobel and laplacian of Gaussian operator. High pass filtering which are not fit for agriculture image noise edge detection because noise and edge belong to the scope of high frequency. In real world applications agriculture images contain object boundaries, object shadows and noise. Therefore, is difficult to distinguish the extract edge from noise, so morphological filter is applied for noise removal. Wavelet Transform Wavelet is multi resolution tool. Wavelet transform have advantages over a Fourier transform, sharp spikes and signal contain discontinuity. In proposed system discrete transform is used. Continues transform is hard to implement and difficult to find out the scaling function. Wavelet transform is used for feature extraction of weed and crop images. 2. Pre-processing Most images have extra parameters such as lighting effects, as well as background information and unnecessary details that may cause misclassification. Therefore, it is important to remove unnecessary parameters for fast and easy processing and to improve the quality of the images. By using the global histogram equalization (GHE) in the pre-processing stage is to improve the quality of the images by lengthening the intensity of the dynamic range and also the histogram of the whole image.