Weed Identification Research Articles

Identification of weeds based on fractal dimension analysis of time series of weed leaf chlorophyll

A new application of fractal dimension analysis was initiated for identification of different kinds of broadleaf weeds. The distribution of chlorophyll time series of weed leaves exhibited self-similar geometricalcharacteristics. The fractal dimension analysis was conducted using Grassberger-Procaccia (G-P) phase space reconstruction algorithm. A total of 300 leaves of three weeds of Oxalis corniculata L. (OC), Ixeris chinensis (Thunb.) Nakai. (IC) and Herba glechomae L. (HG) (species) were sampled and analyzed. Thecorrelation dimensions of time series of O. corniculata, I. chinensis and H. glechomae estimated by G-P algorithm are 8.050, 10.094 and 11.730, respectively. The distribution of chlorophyll time series was restricted in chaos environment and governed by strange attractors. The self-similar distribution property includes the information about weed varieties, which can be used for classification. Key words: Fractal dimension, identification, chlorophylls, weed, Grassberger-procaccia (G-P) algorithm.

Apllication of Computer Technology for Automatic Weeding System

This paper presents the theory of computer automatic weed identification , some methods are discussed in detail including spectrum analysis, color identification, texture assessment and figure analysis methods individually. Besides, the research situation in this field is reviewed and the application trend direction in our country is prospected.

Development and Evaluation of a Prototype Precision Spot Spray System Using Image Analysis to Target Guinea Grass in Sugarcane

Herbicide usage in weed control represents a significant economic cost and environmental risk in Australian sugarcane production. Weed spot spraying has potential to increase sugarcane production while reducing chemical usage and environmentally damaging runoff. However, weed spot spraying is traditionally a laborious manual task. This paper reports on a precision machine vision system that was developed to automatically identify and target the difficult to control weed Panicum spp. (Guinea Grass) in sugarcane crops. The infield machine vision system comprised a camera and artificial illumination to enable day and night trials. Image analysis algorithms were developed to discriminate Guinea Grass and sugarcane based on colour and textural differences between the species. A positive weed identification from the image analysis activated solenoid-controlled spray nozzles. Evaluations of the system in a sugarcane crop established that the image analysis algorithm parameters required frequent recalibration during the day but that the requirement for recalibration was reduced at night with constant artificial illumination. The algorithm was only effective at detecting mature Guinea Grass. The developed technology is considered a viable alternative to manual spot spraying of mature Guinea Grass in sugarcane at night. A cost benefit analysis of the new weed control system indicated potential grower savings of $170/ha by adopting the technology.

THE COMPLETE, TEMPERATURE-RESOLVED EXPERIMENTAL SPECTRUM OF ETHYL CYANIDE (CH3CH2CN) BETWEEN 210 AND 270 GHz

This paper reports the extension of a previously reported experimental method for the identification and characterization of astrophysical weeds in millimeter and submillimeter spectra to the widely used 210–270 GHz atmospheric window. At 300 K, these spectra contain contributions from approximately 40 vibrational states in addition to the cataloged ground state. The quantum mechanical analysis of such a large number of states would be a formidable challenge due to the complex interactions among these dense vibrational states. A new heterodyne receiver-based system is reported, as well as its intensity calibration. Results are presented in the standard astrophysical catalog format as well as our previously described point-by-point format that is effective for the characterization of blends. We also describe and validate an additional spectral synthesis approach, based on the much smaller line list catalog, which is useful in the blended line limit.

An Experimental Ad-Hoc WSN for the Instrumentation of Biological Models

Biological research in agriculture requires a wide range of electronic sensors to fulfill different goals, such as climate monitoring, soil and fruit assessment, control of insects and diseases, chemical pollutants, identification and control of weeds, crop tracking, and so on. This research must be supported by consistent biological models able to simulate diverse environmental conditions to advice on the needed human actions before risky biological damage could become irreversible. In this paper, an experimental agrometeorological sensor-based wireless distributed network is described. Its purpose is to provide real measurements to validate the different biological models used for viticulture applications.

A computer vision approach for weeds identification through Support Vector Machines

This paper outlines an automatic computer vision system for the identification of avena sterilis which is a special weed seed growing in cereal crops. The final goal is to reduce the quantity of herbicide to be sprayed as an important and necessary step for precision agriculture. So, only areas where the presence of weeds is important should be sprayed. The main problems for the identification of this kind of weed are its similar spectral signature with respect the crops and also its irregular distribution in the field. It has been designed a new strategy involving two processes: image segmentation and decision making. The image segmentation combines basic suitable image processing techniques in order to extract cells from the image as the low level units. Each cell is described by two area-based attributes measuring the relations among the crops and weeds. The decision making is based on the Support Vector Machines and determines if a cell must be sprayed. The main findings of this paper are reflected in the combination of the segmentation and the Support Vector Machines decision processes. Another important contribution of this approach is the minimum requirements of the system in terms of memory and computation power if compared with other previous works. The performance of the method is illustrated by comparative analysis against some existing strategies.

ICT and agriculture in India

Information and Communication Technologies (ICTs) have contributed in one way or the other to many transformations in contemporary society. The contexts and the impacts of such transformations are particularly significant for developing societies like India. The relevance of new computing technologies and their effective implementation in developing countries is widely debated, both at policy and community level. Needless to say, FOSS (Free Open Source Software) has encouraged the participation of civil society, creating potential for developing specific information technology tools. In this context, the OSCAR-project (Open Source Simple Computer for Agriculture in Rural Areas) is an initiative from European and South Asian Institutions to assist decision making in agriculture. OSCAR developed a weed identification system for the major weed species in rice-wheat cropping systems of the Indo-Gangetic Plains (IGPs) covering Pakistan, India, Nepal and Bangladesh. An integral part of the OSCAR-project is its applicability to three categories of potential users: farmers, extension officers, and, scientists and students in agricultural sciences. OSCAR was evaluated through extensive interactions with farmer groups, extension personnel, IT specialists, NGO and UN staff, government officials, scientific researchers and PhD students in various disciplines and MSc students from various programs in all the four IGP countries. The experience from OSCAR is helpful in understanding the larger contexts and the impact of ICT interventions in an interdisciplinary framework.

A case study on wireless networks for agricultural applications

Biological research in agriculture needs a lot of specialised electronic sensors in order to fulfil different goals, like as: climate monitoring, soil and fruit assessment, control of insects and diseases, chemical pollutants, identification and control of weeds, crop tracking, and so on. That research must be supported by consistent biological models able to simulate diverse environmental conditions, in order to predict the right human actions before risky biological damage could be irreversible. In this paper an experimental distributed network based on climatic and biological wireless sensors is described, for providing real measurements in order to validate different biological models used for viticulture applications.

Viticulture Zoning by an Experimental WSN

Biological research in agriculture needs a lot of specialized electronic sensors in order to fulfill different goals, like as: climate monitoring, soil and fruit assessment, control of insects and diseases, chemical pollutants, identification and control of weeds, crop tracking, and so on. That research must be supported by consistent biological models able to simulate diverse environmental conditions, in order to predict the right human actions before risky biological damage could be irreversible. In this paper an experimental distributed network based on climatic and biological wireless sensors is described, for providing real measurements in order to validate different biological models used for viticulture applications. Firstly is introduced the rationale of zoning in Galicia’s territory. Then the experimental network for field automatic data acquisition is presented. Following, the design of the wireless network is explained in detail. Finally future developments and conclusions are stated.

Expert system for pests, diseases and weeds identification in olive crops

An expert system was developed with the aim of improving decision-making by olive oil growers. Knowledge was obtained from the literature and from experts. The knowledge was then represented in the knowledge base of the expert system in a series of IF–THEN rules. The system is supported by a data base containing information for the identification of 9 weeds, 14 insects and 14 diseases. The system is enhanced with 150 photos and drawings that assist the used in the identification process. The expert system was evaluated following the conventional expert system evaluation methodologies. According to the validation results the system was considered very satisfactory. The program can be used as an identification tool for farmers and technicians and for educational purposes.

A Real-Time Specific Weed Recognition System Using Statistical Methods

The identification and classification of weeds are of major technical and economical importance in the agricultural industry. To automate these activities, like in shape, color and texture, weed control system is feasible. The goal of this paper is to build a real-time, machine vision weed control system that can detect weed locations. In order to accomplish this objective, a real-time robotic system is developed to identify and locate outdoor plants using machine vision technology and pattern recognition. The algorithm is developed to classify images into broad and narrow class for real-time selective herbicide application. The developed algorithm has been tested on weeds at various locations, which have shown that the algorithm to be very effectiveness in weed identification. Further the results show a very reliable performance on weeds under varying field conditions. The analysis of the results shows over 90 percent classification accuracy over 140 sample images (broad and narrow) with 70 samples from each category of weeds.

Autonomous robotic weed control systems: A review

Autonomous robotic weed control systems hold promise toward the automation of one of agriculture's few remaining unmechanized and drudging tasks, hand weed control. Robotic technology may also provide a means of reducing agriculture's current dependency on herbicides, improving its sustainability and reducing its environmental impact. This review describes the current status of the four core technologies (guidance, detection and identification, precision in-row weed control, and mapping) required for the successful development of a general-purpose robotic system for weed control. Of the four, detection and identification of weeds under the wide range of conditions common to agricultural fields remains the greatest challenge. A few complete robotic weed control systems have demonstrated the potential of the technology in the field. Additional research and development is needed to fully realize this potential.

A vision-based method for weeds identification through the Bayesian decision theory

One of the objectives of precision agriculture is to minimize the volume of herbicides that are applied to the fields through the use of site-specific weed management systems. This paper outlines an automatic computer vision-based approach for the detection and differential spraying of weeds in corn crops. The method is designed for post-emergence herbicide applications where weeds and corn plants display similar spectral signatures and the weeds appear irregularly distributed within the crop's field. The proposed strategy involves two processes: image segmentation and decision making. Image segmentation combines basic suitable image processing techniques in order to extract cells from the image as the low level units. Each cell is described by two area-based measuring relationships between crop and weeds. The decision making determines the cells to be sprayed based on the computation of a posterior probability under a Bayesian framework. The a priori probability in this framework is computed taking into account the dynamic of the physical system (tractor) where the method is embedded. The main contributions of this paper are: (1) the combination of the image segmentation and decision making processes and (2) the decision making itself which exploits a previous knowledge which is mapped as the a priori probability. The performance of the method is illustrated by comparative analysis against some existing strategies.

Weed identification from corn seedling based on computer vision

Weed identification from corn seedling based on computer vision

Research on segmentation of weed images based on computer vision

In this letter, a segment algorithm based on color feature of images is proposed. The algorithm separates the weed area from soil background according to the color eigenvalue, which is obtained by analyzing the color difference between the weeds and background in three color spaces RGB, rgb and HSI. The results of the experiment show that it can get notable effect in segmentation according to the color feature, and the possibility of successful segmentation is 87%–93%. This method can also be widely used in other fields which are complicated in the background of the image and facilely influenced in illumination, such as weed identification, tree species discrimination, fruit picking and so on.

Experiential Learning Activities in the Weed Science Classroom

Considerable discussion has occurred among the weed science community regarding the potential benefits and limitations of integrated approaches to crop and pest management. This discussion also needs to occur in our weed science classrooms, where students from a wide range of academic disciplines are trained in the fundamentals of weed ecology and management. Although the inherent complexity of integrated crop and pest management can make this adaptation to our weed science courses challenging, the use of experiential learning techniques provides an effective means to promote understanding and retention of these concepts. This paper outlines several classroom activities based on the experiential learning approaches that have been implemented by the authors. The activities focus on (1) weed identification and natural history, (2) weed population processes, and (3) integrated management systems. For each activity, we offer our rationale for the exercise, an example of its implementation in the classroom setting, potential pitfalls, and student feedback regarding their perceptions of the activity's educational value. With this paper, we hope to provide examples that may be useful to other weed science educators wishing to incorporate more experiential learning activities into their courses and to initiate a dialogue between educators that can help our community improve and enliven weed science education.

SIMCE: An expert system for seedling weed identification in cereals

Identification of weed seedlings is a difficult task. An expert system to help farmers and extension workers to identify weed species in cereals has been developed. The expert system uses a hierarchical classification and a mix of the text description, photographs and artistic pictures. The system is supported by a data base containing information about 41 weed species and 128 colour images. The expert system was evaluated following the conventional expert system evaluation methodologies. Results of the validation indicated that non-expert users were able to make identification using the expert system. A total of 149 identifications were performed and 63% were identified correctly. The erroneous identifications tended to cluster around monocot species; especially Avena sterilis, Lolium rigidum, Phalaris ssp. and Bromus sterilis were misidentified. Results of the validation process and the writing suggestions provided by the participants were used to implement improvements in the system. The program can be used as an identification tool for farmers and technicians and for educational purposes.

Weed Identification Using Imaging Technology and Spectroscopy

This paper summarizes the authors' research in imaging- and optical weed sensors based on studies of plant spectral and morphological characteristics. Location, color, texture, and shape features of weeds and crops were identified as the major criteria for sensor structure and algorithm design. Imaging-based weed sensors were compared with optical weed sensors. Factors affecting detection accuracy were studied through experiments. Laboratory experiments were also conducted to test the accuracy of statistical and neural-network classification models. Extensive field experiments were conducted to test the effectiveness of the weed sensors.

Site-specific weed management: sensing requirements— what do we need to see?

Automated detection and identification of weeds in crop fields is the greatest obstacle to development of practical site-specific weed management systems. Research progress is summarized for two different approaches to the problem, remote sensing weed mapping and ground-based detection using digital cameras or nonimaging sensors. The general spectral and spatial limitations reported for each type of weed identification system are reviewed. Airborne remote sensing has been successful for detection of distinct weed patches when the patches are dense and uniform and have unique spectral characteristics. Identification of weeds is hampered by spectral mixing in the relatively large pixels (typically larger than 1 by 1 m) and will not be possible from imagery where weed seedlings are sparsely distributed among crop plants. The use of multispectral imaging sensors such as color digital cameras on a ground-based mobile platform shows more promise for weed identification in field crops. Spectral features plus spatial features such as leaf shape and texture and plant organization may be extracted from these images. However, there is a need for research in areas such as artificial lighting, spectral band requirements, image processing, multiple spatial resolution systems, and multiperspective images.

MACHINE VISION FOR ON-LINE WEED IDENTIFICATION

To compensate the existing difficulties in case of machine vision based weed recognition more and more complicated systems are applied. It means a demand of significantly increasing investment and extra cost for the farmers. However, the complexity of the field conditions and the morphological variability of the plants still make weed identification complicated. The typical barrier of the practical application is the insufficient efficiency caused by the limited viewing angle of the optical devices and the long computation time. The authors review their optical sensor based weed monitoring system operating with CCD and infrared camera, and a special solution –a special optical device with an optical angle of 360° – trying to eliminate the limitation of the optical instruments concerning to their view angle.