User:Zaheer1997

Reading Research Paper: Reading the research paper of “Prof. Michelle Araujo E Viegas Mr. Avinash Kurian2 Mr. Victor Joshua Rebello Mr. Niraj Mangaldas Gaunker” and implementing color detection technique and I found good results. A technique of weed detection in crops is described. In this paper, they have developed a method by which we can detect weed using Image processing. By using this method, we can detect weed. They are considering three stages of consideration which will help us to detect the weed easily in the paddy and crop fields. In the first stage, the crop and the weed cannot be separated based on their color and size, so first, they divided the field into rows that separate the area between the crop, and then they can detect weed through pattern recognition. In the second stage, the crop size is greater than the size of the weed so detection can be done based on the size difference between the crop and the weed. This is done by component labeling and also finding aspect ratios. A threshold is defined based on the information and if the value of the aspect ratio or the component labeling exceeds the threshold then it is defined as the weed or it will be considered as the crop. In the third stage, when the crop color turns brown then the weed can be separated easily. Because the weed is mostly green in color. After the detection of weed, the affected area is marked as a white block. The image processing technique has been proved as an effective machine vision system for the agriculture domain. By using this system, you cannot only find weed but also reuse the field for more seeding. There are limitations in this technique which is also described in this paper. These techniques can only be useful for corn and the paddy fields and accuracy varies depending on the algorithms and limitations of image acquisition.