Weed Detection Using Fractal-Based Low Cost Commodity Hardware Raspberry Pi

Abstract

Conventional weed control system is usually used by spraying herbicides uniformly throughout the land. Excessive use of herbicides on an ongoing basis can produce chemical waste that is harmful to plants and soil. The application of precision agriculture farming in the detection IntroductionFarm management systems based on information technology has been widely used to obtain optimum benefit, increase the efficiency of agricultural management [1], protecting the environment [2] and increasing agricultural productivity [3]. Farmers need information from a wide range of ICT tools to identify, analyze, and manage information in spatial and temporal diversity [4] as well as the specific characteristics of the land [5], so that the decision-making process to be more precise during soil preparation, seed selection, fertilizer regulation, management pesticides, watering schedules water and weed management [6].The process of identification of weeds in the field is very important to determine the effective control of this due to lack of proper weed control will cause improper use of herbicides, inefficiencies cost, time and energy [7]. Conventional weed control system is usually done by spraying herbicides uniformly throughout the land [8], it results in excessive use of herbicides will potentially generate waste in the form of chemical residues, emissions to air and soil [9]. Dependence on chemicals also harm human health [10] and the environment [11].The herbicides can be reduced by the application of precision farming application by spraying on right land by detecting weeds on land. Therefore, precision farming is needed to determine the level of weed vegetation in order to control the conditions and needs of the plant based on the specific characteristics of the land [12]. Precision farming is the application of information technology in agricultural management systems that allow rigorous treatment (precise treatment) agribusiness chain from upstream (on farm) to downstream (off farm) [13].Computer vision as one of the precision farming applications is very promising [14] which can be used for the identification and classification of plants. OpenCV is a library Public License can be used to detect the image of weeds. Weed detection in realtime is still difficult to implement in the field due to need a large place and the use of large electric power. The need for specification of minicomputer and small power consumption has attracted the attention of

Development of a Light-Weight Unmanned Aerial Vehicle for Precision Agriculture

“…This is a credit-card-like single-board computer that uses the open-source Linux operating system and is used to provide access to the internet and connect automatic systems [ 32 ]. It was initially developed by the Raspberry Pi foundation in the United Kingdom to assist students in learning basic programming skills.…”

Optimization of Parallel K-means for Java Paddy Mapping Using Time-series Satelite Imagery

“…The improvement of energy efficiency not only depending on the problem size but also the degree of multi-threading [22]. Low-power consumption computer such as Raspberry Pi (3.5 Watt) can perform as a desktop pc and used for image processing and weed fractal dimension processing [23].…”

Implementation of Robust Tracking Algorithm on Nano-Computer

“…The system can be used for the fractal size treatment of weeds, the average speed ratio between the PC and Raspberry Pi is 0.04 times faster. The authors deduce that the use of Raspberry Pi is cheaper and the energy consumption is efficient compared to a personal computer, by [6]. SIFT algorithm for palmar vein recognition is proposed is built on Raspberry Pi.…”