A computer vision approach for weeds identification through Support Vector Machines

This paper outlines an automatic computervision 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 SupportVectorMachines and determines if a cell must be sprayed. The main findings of this paper are reflected in the combination of the segmentation and the SupportVectorMachines 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.

Automatic detection of crop rows in maize fields with high weeds pressure

This paper proposes a new method, oriented to crop row detection in images from maize fields with high weed pressure. The vision system is designed to be installed onboard a mobile agricultural vehicle, i.e. submitted to gyros, vibrations and undesired movements. The images are captured under image perspective, being affected by the above undesired effects. The image processing consists of three main processes: image segmentation, double thresholding, based on the Otsu’s method, and crop row detection. Image segmentation is based on the application of a vegetation index, the double thresholding achieves the separation between weeds and crops and the crop row detection applies least squares linear regression for line adjustment. Crop and weed separation becomes effective and the crop row detection can be favorably compared against the classical approach based on the Hough transform. Both gain effectiveness and accuracy thanks to the double thresholding that makes the main finding of the paper.

Discriminating Crop, Weeds and Soil Surface with a Terrestrial LIDAR Sensor

In this study, the evaluation of the accuracy and performance of a light detection and ranging (LIDAR) sensor for vegetation using distance and reflection measurements aiming to detect and discriminate maize plants and weeds from soil surface was done. The study continues a previous work carried out in a maize field in Spain with a LIDAR sensor using exclusively one index, the height profile. The current system uses a combination of the two mentioned indexes. The experiment was carried out in a maize field at growth stage 12–14, at 16 different locations selected to represent the widest possible density of three weeds: Echinochloa crus-galli (L.) P.Beauv., Lamium purpureum L., Galium aparine L.and Veronica persica Poir.. A terrestrial LIDAR sensor was mounted on a tripod pointing to the inter-row area, with its horizontal axis and the field of view pointing vertically downwards to the ground, scanning a vertical plane with the potential presence of vegetation. Immediately after the LIDAR data acquisition (distances and reflection measurements), actual heights of plants were estimated using an appropriate methodology. For that purpose, digital images were taken of each sampled area. Data showed a high correlation between LIDAR measured height and actual plant heights (R 2 = 0.75). Binary logistic regression between weed presence/absence and the sensor readings (LIDAR height and reflection values) was used to validate the accuracy of the sensor. This permitted the discrimination of vegetation from the ground with an accuracy of up to 95%. In addition, a Canonical Discrimination Analysis (CDA) was able to discriminate mostly between soil and vegetation and, to a far lesser extent, between crop and weeds. The studied methodology arises as a good system for weed detection, which in combination with other principles, such as vision-based technologies, could improve the efficiency and accuracy of herbicide spraying.

Automatic segmentation of relevant textures in agricultural images

One important issue emerging strongly in agriculture is related with the automatization of tasks, where the optical sensors play an important role. They provide images that must be conveniently processed. The most relevantimage processing procedures require the identification of green plants, in our experiments they come from barley and corn crops including weeds, so that some types of action can be carried out, including site-specific treatments with chemical products or mechanical manipulations. Also the identification of textures belonging to the soil could be useful to know some variables, such as humidity, smoothness or any others. Finally, from the point of view of the autonomous robot navigation, where the robot is equipped with the imaging system, some times it is convenient to know not only the soil information and the plants growing in the soil but also additional information supplied by global references based on specific areas. This implies that the images to be processed contain textures of three main types to be identified: green plants, soil and sky if any. This paper proposes a new automatic approach for segmenting these main textures and also to refine the identification of sub-textures inside the main ones. Concerning the green identification, we propose a new approach that exploits the performance of existing strategies by combining them. The combination takes into account the relevance of the information provided by each strategy based on the intensity variability. This makes an important contribution. The combination of thresholding approaches, for segmenting the soil and the sky, makes the second contribution; finally the adjusting of the supervised fuzzy clustering approach for identifying sub-textures automatically, makes the third finding. The performance of the method allows to verify its viability for automatic tasks in agriculture based on image processing

Mapping wide row crops with video sequences acquired from a tractor moving at treatment speed

This paper presents a mapping method for wide row crop fields. The resulting map shows the crop rows and weeds present in the inter-row spacing. Because field videos are acquired with a camera mounted on top of an agricultural vehicle, a method for image sequence stabilization was needed and consequently designed and developed. The proposed stabilization method uses the centers of some crop rows in the image sequence as features to be tracked, which compensates for the lateral movement (sway) of the camera and leaves the pitch unchanged. A region of interest is selected using the tracked features, and an inverse perspective technique transforms the selected region into a bird’s-eye view that is centered on the image and that enables map generation. The algorithm developed has been tested on several video sequences of different fields recorded at different times and under different lighting conditions, with good initial results. Indeed, lateral displacements of up to 66% of the inter-row spacing were suppressed through the stabilization process, and crop rows in the resulting maps appear straight

Real-time image processing for crop/weed discrimination in maize fields

This paper presents a computer vision system that successfully discriminates between weed patches and crop rows under uncontrolled lighting in real-time. The system consists of two independent subsystems, a fast image processing delivering results in real-time (Fast Image Processing, FIP), and a slower and more accurate processing (Robust Crop Row Detection, RCRD) that is used to correct the first subsystem's mistakes. This combination produces a system that achieves very good results under a wide variety of conditions. Tested on several maize videos taken of different fields and during different years, the system successfully detects an average of 95% of weeds and 80% of crops under different illumination, soil humidity and weed/crop growth conditions. Moreover, the system has been shown to produce acceptable results even under very difficult conditions, such as in the presence of dramatic sowing errors or abrupt camera movements. The computer vision system has been developed for integration into a treatment system because the ideal setup for any weed sprayer system would include a tool that could provide information on the weeds and crops present at each point in real-time, while the tractor mounting the spraying bar is moving

Seed image analysis and taxonomy of Diplotaxis DC. (Brassicaceae, Brassiceae)

The genus Diplotaxis, comprising 32 or 34 species, plus several additional infraspecific taxa, displays a considerable degree of heterogeneity in the morphology, molecular markers, chromosome numbers and geographical amplitude of the species. The taxonomic relationships within the genus Diplotaxis were investigated by phenetic characterisation of germplasm belonging to 27 taxa of the genus, because there is an increasing interest in Diplotaxis, since some of its species (D. tenuifolia, D. muralis) are gathered or cultivated for human consumption, whereas others are frequent arable weeds (D. erucoides) in many European vineyards. Using a computer-aided vision system, 33 morpho-colorimetric features of seeds were electronically measured. The data were used to implement a statistical classifier, which is able to discriminate the taxa within the genus Diplotaxis, in order to compare the resulting species grouping with the current infrageneric systematics of this genus. Despite the high heterogeneity of the samples, due to the great intra-population variability, the stepwise Linear Discriminant Analysis method, applied to distinguish the groups, was able to reach over 80% correct identification. The results obtained allowed us to confirm the current taxonomic position of most taxa and suggested the taxonomic position of others for reconsideration.

Search strategies and the automated control of plant diseases

We propose the use of the "infotaxis" search strategy as the navigation system of a robotic platform, able to search and localize infectious foci by detecting the changes in the profile of volatile organic compounds emitted by and infected plant. We builded a simple and cost effective robot platform that substitutes odour sensors in favour of light sensors and study their robustness and performance under non ideal conditions such as the exitence of obstacles due to land topology or weeds.

Crop row detection in maize fields inspired on the human visual perception

This paper proposes a new method, oriented to image real-time processing, for identifying crop rows in maize fields in the images. The vision system is designed to be installed onboard a mobile agricultural vehicle, that is, submitted to gyros, vibrations, and undesired movements. The images are captured under image perspective, being affected by the above undesired effects. The image processing consists of two main processes: image segmentation and crop row detection. The first one applies a threshold to separate green plants or pixels (crops and weeds) from the rest (soil, stones, and others). It is based on a fuzzy clustering process, which allows obtaining the threshold to be applied during the normal operation process. The crop row detection applies a method based on image perspective projection that searches for maximum accumulation of segmented green pixels along straight alignments. They determine the expected crop lines in the images. The method is robust enough to work under the above-mentioned undesired effects. It is favorably compared against the well-tested Hough transformation for line detection.

Even-sowing pattern strategies for a low-input agricultural system in forage maize

The objective of this study was to verify the effectiveness of new patterns of sowing and to achieve a low-input organic system in two different environments (northern and southern Europe). The study was motivated by the hypothesis that more even sowing patterns (triangular and square) would significantly enhance the growth and yield of forage maize under widely varying conditions, compared with traditional mechanised rectangular seed patterns. An experiment was conducted in Madrid and duplicated in Copenhagen during 2010. A random block design was used with a 2 × 2 factorial arrangement based on two seed-sowing patterns: traditional (rectangular) and new (even) and two weed-management conditions (herbicide use and a low-input system). In both weed-management conditions and locations, the production of aerial maize biomass was greater for the new square seed patterns. In addition, the new pattern showed a greater effectiveness in the control of weeds, both at the initial crop stages (36 and 33% fewer weeds m-2 at the 4- and 8-leaf stages, respectively, in the Copenhagen field experiment) and at the final stage. The final weed biomass for the new pattern was 568 kg ha-1 lower for the Copenhagen experiment and 277 kg ha-1 lower in Madrid field experiments. In the light of these results, the new pattern could potentially reduce the use of herbicides. The results of the experiments support the hypothesis formulated at the beginning of this study that even-sowing patterns would be relatively favourable for the growth and yield of the maize crop. In the near future, new machinery could be used to achieve new seed patterns for the optimisation of biomass yield under low-input systems. This approach is effective because it promotes natural crop-weed competition.

Automatic expert system based on images for accuracy crop row detection in maize fields

This paper proposes an automatic expert system for accuracy crop row detection in maize fields based on images acquired from a vision system. Different applications in maize, particularly those based on site specific treatments, require the identification of the crop rows. The vision system is designed with a defined geometry and installed onboard a mobile agricultural vehicle, i.e. submitted to vibrations, gyros or uncontrolled movements. Crop rows can be estimated by applying geometrical parameters under image perspective projection. Because of the above undesired effects, most often, the estimation results inaccurate as compared to the real crop rows. The proposed expert system exploits the human knowledge which is mapped into two modules based on image processing techniques. The first one is intended for separating green plants (crops and weeds) from the rest (soil, stones and others). The second one is based on the system geometry where the expected crop lines are mapped onto the image and then a correction is applied through the well-tested and robust Theil–Sen estimator in order to adjust them to the real ones. Its performance is favorably compared against the classical Pearson product–moment correlation coefficient.

Evaluación de tecnologías de manejo de arvenses en el cultivo de la caña de azúcar (Saccharum spp. híbrido)

En Cuba, las arvenses que afectan al cultivo de la caña de azúcar son una de las causas fundamentales de los bajos rendimientos agrícolas y su control constituye unas de las principales partidas de gastos. En general, se aplican los herbicidas, así como otros métodos de control, sin tener en cuenta el tipo de suelo y las características de estas plantas. Sobre el manejo de arvenses no existen trabajos de investigación que aborden aspectos de eficiencia energética de las producciones y daños al ambiente. Por lo antes señalado, el objetivo de esta investigación fue evaluar diversas tecnologías de manejo de arvenses en el cultivo de la caña de azúcar (Saccharum spp. híbrido), en cepas de primavera y retoño, en tres tipos de suelos, con el propósito de obtener producciones sustentables. El área de estudio se localizó en los campos de la Empresa Azucarera del municipio “Majibacoa”, provincia de Las Tunas (oriente de Cuba), que posee condiciones edafoclimáticas que abundan a lo largo del país. Los tres tipos de suelos más representativos son Fersialítico Pardo Rojizo ócrico, Pardo Mullido y Vertisol Crómico gléyco. En dicha área se han identificado 31 especies de arvenses, 16 de la clase Liliopsida y 15 de la Magnoliopsida. En un primer grupo de experimentos, se desarrollaron nueve ensayos de campo para evaluar la efectividad de herbicidas y mezclas de estos en el manejo de arvenses en el cultivo de la caña de azúcar, tanto en cepas de primavera como de retoño, en los tres tipos de suelos. Se establecieron parcelas de 80 m2 distribuidas en bloques al azar con cuatro réplicas. La efectividad se evaluó por medio del porcentaje de cobertura por arvenses y la fitotoxicidad provocada a las plantas de caña, teniendo en cuenta el coste asociado a cada tratamiento. En aplicaciones preemergentes en caña planta de primavera, el herbicida más eficiente fue el Isoxaflutole con dosis de 0,15; 0,20 y 0,25 kg.ha-1 de producto comercial (pc) en los suelos Fersialítico, Pardo y Vertisol respectivamente. En aplicaciones postemergentes tempranas la mezcla más eficiente fue la de Isoxaflutole + Ametrina +2,4-D con las dosis de Isoxaflutole citadas anteriormente. En aplicaciones preemergentes en cepa de retoño, el herbicida más eficiente fue el Isoxaflutole a dosis de 0,20 kg.ha-1 pc para el suelo Fersialítico y a 0,25 kg.ha-1 pc para los suelos Pardo y Vertisol. En un segundo grupo, se realizaron seis ensayos de campo distribuidos en dos fases. En la primera fase, se desarrollaron tres experimentos, uno por cada tipo de suelo, para evaluar la eficiencia de nueve tecnologías de manejo de arvenses (químicas y físicas combinadas) en cepa de primavera de caña de azúcar. En la siguiente fase, los tres ensayos restantes (uno por tipo de suelo) evaluaron tecnologías de manejo de arvenses durante dos ciclos de producción de caña de azúcar (etapa de primavera y retoño). En la etapa de primavera se aplicó la tecnología más eficiente de los tres experimentos anteriores y durante la etapa de retoño se evaluaron otras nueve tecnologías propias de este tipo de cepa. En estos experimentos los diferentes tratamientos se aplicaron en franjas distribuidas al azar con cuatro réplicas. En las tecnologías evaluadas se emplearon los herbicidas y mezclas que resultaron más eficientes en el primer grupo de experimentos. En cada caso, se evaluaron la eficiencia energética de la producción de azúcar y otros derivados, la resistencia a la penetración de los suelos, la carga contaminante hacia la atmósfera producto de la combustión del diésel y los beneficios al aplicar las diferentes tecnologías. En la primera fase (cepa de primavera), la tecnología con mejor resultado fue la aplicación preemergente de Isoxaflutole inmediatamente después de la plantación, seguida de descepe químico con Glufosinato de amonio, más labor con grada múltiple aproximadamente a los 80 días de la plantación y aplicación pre-cierre con Glufosinato de amonio. En la segunda fase (dos ciclos del cultivo), el mejor resultado se obtuvo cuando en la etapa de retoño se realizó una aplicación preemergente de Isoxaflutole, descepe químico con Glufosinato de amonio y aplicación pre-cierre con este mismo herbicida. En los tres tipos de suelos durante los dos ciclos, la eficiencia energética tuvo valores de 7,2 - 7,5, la resistencia a la penetración 1,2 - 1,5 MPa, la carga contaminante hacia la atmósfera fue de 63,3 - 64,9 kg.t-1 de caña cosechada y beneficios de 8.324 - 8.455 pesos cubanos por hectárea. Este estudio demuestra que un control eficiente de las arvenses debe tener en cuenta necesariamente el tipo de suelo. Así, en los Vertisoles, con mayor contenido en arcilla, se requieren mayores dosis de Isoxaflutole y la eficiencia energética de la producción es menor. La persistencia de ciertas arvenses, especialmente de la clase Liliopsida, requiere de un manejo integrado que incluya diferentes tipos de herbicidas. ABSTRACT In Cuba, weeds affecting the sugarcane are one of the main causes of low agricultural yields, and their control constitutes some of the main items of expenditure. In general, herbicides are applied, as well as other control methods, without keeping in mind the soil type and the characteristics of these plants. Moreover, weed control research approaching aspects about energy efficiency of the crop production, and environmental damages are missing. Hence, the objective of this investigation was to evaluate diverse technologies of weed handling in sugarcane (Saccharum spp. hybrid), both in spring cane plant and ratoon, in three types of soils, with the purpose of obtaining sustainable productions. The study area was located in the fields of the Sugar Enterprise of the Municipality "Majibacoa”, Las Tunas province (east of Cuba) that possesses ecological conditions that are plentiful along the country. The three more representative types of soils are Fersialitic, Brown, and Vertisol. In this area 31 weeds species have been identified, 16 of the Class Liliopsida and 15 of the Magnoliopsida. In a first group of experiments, nine field rehearsals were developed to evaluate the effectiveness of herbicides and mixtures of these for weed handling in sugarcane, in spring cane plants as well as in ratoon, in the three types of soils. Plots of 80 m-2 were distributed at random blocks with four replications. The effectiveness was evaluated by means of the covering percentage by weeds and the provoked toxicity to the cane plants, keeping in mind the cost associated to each treatment. In preemergence applications in spring cane plant, the most efficient herbicide was the Isoxaflutole with dose of 0.15; 0.20 and 0.25 kg.ha-1 of commercial product (pc) in the soils Fersialítico, Brown and Vertisol respectively. In early postemergence applications the most efficient mixture was that of Isoxaflutole + Ametrina + 2,4-D with the doses of Isoxaflutole mentioned previously. In preemergence applications in ratoon, the most efficient herbicide was the Isoxaflutole at dose of 0.20 kg.ha-1 pc for the soil Fersialític and to 0.25 kg. ha-1 pc for the Brown soil and Vertisol. In a second group, six field rehearsals distributed in two phases were carried out. In the first phase, three experiments were developed, one for each soil type, to evaluate the efficiency of nine technologies of weed handling in spring cane plant. In the following phase, the three remaining rehearsals (one for each soil type) diverse technologies of weed handling were evaluated during two cycles of sugarcane production (spring stage and ratoon). In the spring stage the most efficient technology in the three previous experiments was applied and during ratoon stage other nine technologies were evaluated. In these experiments the different treatments were applied in fringes distributed at random with four replicas. In the evaluated technologies the herbicides and mixtures were used selecting those that were more efficient in the first group of experiments. In each case, the energy efficiency of the sugar production and other derivatives, the soil penetration resistance, the polluting load toward the atmosphere product of the combustion, and the benefits when applying the different technologies were all evaluated. In the first phase (spring cane plant), the technology with better result was the preemergence application of Isoxaflutole immediately after the plantation, followed by chemical eradication with Ammonia Glufosinate, hoeing work with multiple tier approximately to the 80 days of the plantation and pre-closing application with Ammonia Glufosinate. In the second phase (two cycles of the cultivation), the best result was obtained when a preemergence application of Isoxaflutole was carried out in sprout's stage, chemical eradication with Ammonia Glufosinate and pre-closing application with this same herbicide. In the three types of soils during the two cycles, the energy efficiency achieved values of 7.2 to 7.5, the resistance to the penetration 1.2 - 1.5 MPa, the polluting load toward the atmosphere was of 63.3 - 64.9 kg.t-1 of the harvested cane and the obtained benefits of 8,324 - 8,455 Cuban pesos per hectare. This study demonstrates that an efficient control of the weeds should necessarily keep in mind the soil type. This way, in the Vertisols, with more clay content, bigger dose of Isoxaflutole is required and the energy efficiency of the production is smaller. The persistence of certain weeds, especially of the class Liliopsida, requires of an integrated handling him to include different types of herbicides.