- Eurasian Journal of Agricultural Research
- Vol: 1 Issue: 1
- A Software Development for Real Time Spray Control System in Herbicide Application
A Software Development for Real Time Spray Control System in Herbicide Application
Authors : Ömer Barış Özlüoymak, Ali Bolat, Ali Bayat, Emin Güzel
Pages : 13-19
View : 22 | Download : 4
Publication Date : 2017-08-30
Article Type : Research
Abstract :Advances in different technologies, such as high-resolution vision systems, innovative sensors and embedded computing systems, are finding direct application in agriculture. In precision farming, image analysis techniques can aid farmers in herbicide applications, and thus lower the risk of soil and water pollution by reducing the amount of chemicals applied. Optical sensors and computer vision, which can be used in automated weed detection and control spray systems, are being used in recent years extensively. A real-time auto tracking and determination system for weed detection and spray on/off were designed, built and set up in the laboratory at the Department of Agricultural Machinery and Technologies Engineering of Çukurova University. In this study; to get the target images, a web camera, mounted at a height of 50 cm above the target object was used. During the start of the weed tracking operation, the web camera captured images of the artificial weeds. Developed software, which could be reprogrammed and adjusted according to the user preference, was created by using LabVIEW. Weed coverage was determined from each image by using a "greenness method” in which the red, green, and blue intensities of each pixel were compared. The sprayer nozzle was turned ‘on’ or ‘off’ by using a data acquisition card and a relay card, depending on the green color pixels of weeds. The sprayer valve opened the nozzle when the camera detected the presence of weeds. Image processing performance of this system, in where nozzle and camera were mounted at a stationary position while weeds were on a movable belt, was tested at the different speeds of conveyor belt consisted of an inverter drive system and 3 phase 4 pole electric motor. The laboratory performance evolution revealed that the system could detect the weeds successfully and could be used to decrease the herbicide quantity.Keywords : Image processing, LabVIEW, Machine vision, Weed detection