Mechanical Weed Research Articles

A peanut and weed detection model used in fields based on BEM-YOLOv7-tiny.

Due to the different weed characteristics in peanut fields at different weeding periods, there is an urgent need to study a general model of peanut and weed detection and identification applicable to different weeding periods in order to adapt to the development of mechanical intelligent weeding in fields. To this end, we propose a BEM-YOLOv7-tiny target detection model for peanuts and weeds identification and localization at different weeding periods to achieve mechanical intelligent weeding in peanut fields at different weeding periods. The ECA and MHSA modules were used to enhance the extraction of target features and the focus on predicted targets, respectively, the BiFPN module was used to enhance the feature transfer between network layers, and the SIoU loss function was used to increase the convergence speed and efficiency of model training and to improve the detection performance of the model in the field. The experimental results showed that the precision, recall, mAP and F1 values of the BEM-YOLOv7-tiny model were improved by 1.6%, 4.9%, 4.4% and 3.2% for weed targets and 1.0%, 2.4%, 2.2% and 1.7% for all targets compared with the original YOLOv7-tiny. The experimental results of positioning error show that the peanut positioning offset error detected by BEM-YOLOv7-tiny is less than 16 pixels, and the detection speed is 33.8 f/s, which meets the requirements of real-time seedling grass detection and positioning in the field. It provides preliminary technical support for intelligent mechanical weeding in peanut fields at different stages.

A green and economical weeds control agent suitable for drone

Weeds are one of the main factors affecting orchard yield, fruit quality and economic benefits. However, present methods have been lower effect and high labor intensity. To inhance the effect of weeds control on orchard, a new green and economic agent suitable for drone with choline chloride and eucalyptus oil as the main components was developed. Adopting optimized “3414” test design, the paper studied the prevention and control effect of this new weed control agent on 8 kinds of main weeds in the orchards of the Yellow-Huai Rivers. The results showed that T5 and T12 were significantly (P≤0.05) control the growth speed of 8 kinds of weeds, T4 was only inhibit the growth of Erigeron canadensis, Digitaria sanguinalis and Artemisia annua, T11 was only inhibit the growth of Tribulus terrestris, Humulus scandens and Cyperus rotundus, and T3 and T13 were promote the growth of weeds, which had the worst control effect among all treatments. Combined with the economic benefit analysis, the cost per 667m2 of weed control of T12 was 124 RMB, which was the most economical compared with manual weeding and mechanical weeding. Therefore, applying 30g choline chloride, 120 mL eucalyptus oil and 75g sucrose ester to each 667m2 is recommended to be popularized as a new green and economical weed control agent in the orchard production of the Yellow-Huai Rivers.

Comparing manual and mechanical weed management techniques for upland organic rice in acidic soil of Meghalaya: A on-farm study

Comparing manual and mechanical weed management techniques for upland organic rice in acidic soil of Meghalaya: A on-farm study

Seletividade de herbicidas e controle de plantas daninhas no cultivo de gladíolo

Abstract The objective of this work was to evaluate the selectivity of pre- and post-emergent herbicides and their efficiency in weed control, as well as the growth and quality of floral stems, in a gladiolus (Gladiolus x grandiflorus) crop in two growing seasons. The experiments were carried out in the field, where the planted gladiolus received the application of pre- and post-emergent herbicides, with and without mechanical weed control. At 7, 14, 21, and 28 days after emergence (DAE), monocotyledon and eudicotyledon control and phytotoxicity to gladiolus plants were evaluated. At 28 DAE, the shoot dry matter of weeds and gladiolus plants was also evaluated. The s-metolachlor and diuron pre-emergent herbicides and the haloxyfop-p-methyl post-emergent herbicide are selective to gladiolus plants in both seasons, which allows of the production of floral stems with commercial quality. The s-metolachlor and haloxyfop-p-methyl herbicides also promote an efficient control of monocotyledons regardless of the growing season. Although, when applied, saflufenacil and 2,4-D cause phytotoxicity in the vegetative phase of gladiolus, they do not affect the production of floral stems and control satisfactorily eudicotyledons in the second season.

Različite kontrole mehaničkog načina uništavanja korova u vinogradu zbog redukcije ekološkog otiska

The purpose of the study was to investigate the alternative methods of weed control in the vineyard under 'Sauvignon' vines to reduce ecological footprint in vineyard production. Weed control was performed at the experimental site with the mechanical weed control under vines with a grass cutter, a rotating star hoe (roll hoe) and a vine trunk cleaner, while at the same time additional mulching was carried out in the belts between the rows. The ecological footprint of each production operation was estimated by including environmental impacts related to fossil-C (kg CO2/hm2 ), air, water, soil, nonrenewable, renewable, and area resources. While carrying out the mechanical control, we analysed the weed coverage, grape yield and yield loss. Grape yield was highest for weedfree control plots, followed by plots where ploughshare was used. Yield loss was lowest when using ploughshare. In addition to the weed-free plot, weed coverage was lowest when glyphosate was used. The maximum amount of CO2 is released while using rotating star hoe. At the grass cutter and vine trunk cleaner less CO2 was released. The same is with global warming potential (GWP).

Effect of organic cultivation technology on row crop yielding capacity

The study aimed to estimate the effect of organic technologies on yielding capacity of row crops in the North-West Russia and the after-effect of compost on the successor crop. Experiments took place in 2020-2022. The crop rotation included the zoned perennial grasses, potato, and table beet. Potatoes were fertilised with BIAGUM compost produced from poultry manure in a fermenter. In 2020-2021, the compost application doses were 0 (control), 4 and 8 t ha-1 (5.5 t ha-1 in 2022) or 0 (control), 80 and 160 kg N ha-1 (110 kg N ha-1 in 2022). The table beet was sown on the plots fertilised in the previous year under potatoes. The mechanical weed control of row spacing was performed twice in the growing season. Weather conditions were monitored by the own weather station near the experimental field. The year 2021 was the warmest and the driest in the study period. In 2022, the rainfall was satisfactory for crops – 238.20 mm. In control variants, the potato yields ranged from 12.78 to 20.25 t ha-1 that was 25.6-67.5% of the cultivated variety potential. Application of 4 t ha-1 of compost only slight affected the yield. But with higher compost application dose, a significant yield gain (greatest 56.88%) was observed, especially under favourable climatic conditions of 2020. The biggest beetroot average yield – from 22.5 to 27.7 t ha-1 – was also in 2020. In 2022, the average yield was from 21.5 to 24.3 t ha-1. By the study results, the applied organic technologies proved effective for producing potatoes and table beet in the North-West Russia.

Effect of Sowing Methods and Irrigation Levels on wheat (Triticum aestivum L.)

The field experiment was carried out at agricultural research farm of IFTM University Moradabad (U.P.), India during rabi season 2018-19 to study the effect of sowing methods and irrigation levels on wheat. Sowing method and irrigation being a major constraint to reduces the potential of wheat yield. The treatments were arranged in split plot design in three replications, sowing methods in main plots and irrigation levels in sub-plots. The experiment treatments included two sowing methods, i.e. [S1 – conventional sowing method and S2 - furrow irrigated raised bed (FIRB) system] in main plots and six irrigation levels [I0 – (control), I1 – (one irrigation at CRI stage), I2 – (two irrigation at CRI and tillering stage), I3 – (three irrigations at CRI, tillering and late jointing stage), I4 – (four irrigations at CRI, tillering, late jointing and milking stage) and I5 – (five irrigations at CRI, tillering, late jointing, milking and dough stage). Wheat crop was sown by traditional method and furrow irrigated raised beds (FIRB) method. Raised beds were accommodating 3 rows of wheat at 20 cm wide. Furrows that are in between the beds are used for irrigation. This system permits adequate saving of irrigation water. Results indicated that sowing of wheat on FIRB system surpassed the conventional method of sowing for plant height, number of tillers plant-1, dry weight (g plant-1), number of effective tillers plant-1, spike length, number of grain spike-1, 1000 grain weight, grain yield, straw yield and biological yield followed by conventional sowing method. On the other hand irrigation levels I5 - (five irrigations at CRI, tillering, late jointing, milking and doughing stage) recorded the highest values for most of growth, yield components and yield of wheat. In FIRB system beds gave the opportunity for mechanical weeding and fertilizer placement and also improved the fertilizer and water use efficiencies. The role of sowing methods and level of irrigation in plant health and growth has been investigated in crop and it seemed significantly affecting with treatment.

Communication between agricultural robot and mechanical weeding machine based on ISO 11783 network

Latest technologies such as advanced agriculture implements and autonomous vehicles designed with the latest sensor technologies could partially offer a solution to the challenges in agriculture that are related to crop demands, increased cost of inputs, and labor shortage. However, the compatibility, in terms of communication between the autonomous vehicle and the agricultural implements, raises the main concern. This is because most of the autonomous robotic platforms in agriculture are not compliant with existing agricultural implements that are integrated with ISO 11783 (also known as ISOBUS) standards. Besides that, there are still few agricultural processes where the implementation of ISOBUS would fulfill farmers’ needs to evaluate the performance of the implements. Weeding is one of those processes as it has a significant influence on crop growth. To address the mentioned issues, this work aims at integrating an agricultural robot with an existing mechanical weeder by leveraging ISOBUS in combination with software and hardware level.To fulfill the aim, this paper outlines the development of a middleware that makes communication between the mobile robot and the weeder possible. In addition to that, the development of an implement object pool (IOP) for the representation of the weeder performance, in terms of weeding quality, and its integration with the simulated electronic control unit (ECU) of the weeder is discussed. A preliminary analysis and assessment defined the threshold of 10% for the weeding quality that raised the STOP flag for the application.

Alternate of Manual Weeding Tools: A Research into an Automatic Weeding Control Strategies Enabled by Embedded Systems

A key enabling factor for treating individualized weeds in the crops is the Automatic Weeding Control device (AWCd) as it is being capable of finding and distinguishing weeds in the field. The non-herbicide approaches used to manage weeds are a component of the personalized treatment of weed crops. In this study, mechanical weed control techniques are investigated as an alternative. In this study, three specifications—a cutting mechanism, a head-surface tilling (tines), and a foot-surface tilling (arrow hoe)—are taken into account. With different rates of application to herbicide-resistant, these processes were estimated in a controlled field and the efficacy of the implementation is estimated by using the demographic analysis and also, the importance of early interface has also been highlighted. For all the weeds accounted in this research, Automatic weeding control was found to be very effective thus ended up in overall survival probability.

Study of Mechanical-Chemical Synergistic Weeding on Characterization of Weed–Soil Complex and Weed Control Efficacy

Mechanical-chemical synergy has been proven efficient in weed control. However, characterizing the state of the weed–soil complex after mechanical weeding and revealing its effects on subsequent herbicide application is still challenging, which restricts the implementation of this technology. This paper first presents a method to characterize the state of the weed–soil complex from the perspectives of the fragmentation and composite characteristics. The regrowth of the weed–soil complex and the effects of complemented herbicide-reduced spraying on weed control efficacy and crop yield were then investigated. The results showed that the typical diameters of the weed–soil complexes were 10.67 cm and 2.82 cm after inter-row hoe shovel and intra-row finger weeding, respectively. There were mainly two and four weed–soil complex states after inter-row and intra-row weeding, respectively. The regrowth rate corresponding to the weed–soil complex state with the largest component proportion after inter-row and intra-row weeding was 76.91% and 18.37%, respectively. The additional chemical herbicide sprayed on the weed–soil complex significantly improved the fresh weight control efficacy of 95.12% for the preposed inter-row mechanical weeding and 138.07% for the preposed intra-row mechanical weeding in the maize silking stage. The maize yield of inter-row mechanical–75% chemical application treatment was 9.27% higher than that of chemical treatment. Mechanical weeding creates a suitable weed–soil complex state for subsequent chemical application and improves the synergistic weeding effect.

Citizen support for agri-environmental measures motivated by environmental consciousness

Citizens are indispensable stakeholders in rural social-ecological systems and public policies that affect them. For example, the EU’s Common Agricultural Policy (CAP) promotes agri-environmental measures (AEMs) to reduce agricultural externalities and enhance supply of ecosystem services to benefit society, compensating farmers for incurred costs and income foregone with public money. However, studies on citizen support for AEMs are rather rare. This study assesses citizen preferences for specific AEMs with a discrete choice experiment in a rural area in Flanders, which stands as an example for densely populated, intensive agricultural landscapes. Hybrid choice modelling based on the multidimensional environmental consciousness framework revealed support and significant willingness to pay for all AEMs, although with substantial heterogeneity. Mechanical weeding and increasing nesting opportunities for bees and birds were particularly favored. The affective dimension of environmental consciousness, i.e., environmental attitude, only plays a limited role in shaping the observed preferences. The cognitive dimension is even less relevant for specific preferences. This should be considered if one aims to realize community supported policies. The study presents the applicability of hybrid choice modelling in the field of environmental economics and highlights some methodological issues.

Biomass-powered thermal weeding in wine farms: An environmental and economic assessment

The energy transition in agriculture must be accompanied by agronomic practices that shift production towards organic regimes. Among the energy consuming processes that today find resistance in organic conversion there is certainly the removal of weeds from crops, that is mainly carried out through herbicides. This analysis evaluates the economic and environmental implications that the introduction of innovative thermal weeding systems, powered by biomass gasification, can have in the substitution of traditional chemical and mechanical weeding technologies. The comparison is carried out through a combination of cost–benefit analysis, environmental analysis and sensitivity analysis applied to the case study of vineyard management in northern Italy. The results show that the biomass-powered weeding system is economically feasible when it can benefit from economic incentives that reward organic production. In all cases, biomass-powered weeding system provides for a reduction in equivalent CO2 emissions which reaches −75% in the case of replacement of glyphosate-based chemical weeding methods. The analysis points out that gasification-powered weeding process can also achieve carbon neutrality through the co-production of biochar, which should lead policy makers to support this practice under the environmental objectives at national and European level.

The Positive Effects of Mechanical and Chemical Treatments with the Application of Biostimulants in the Cultivation of Solanum tuberosum L.

The aim of the study was to determine the effect of mechanical and chemical treatments with the application of biostimulants on the effectiveness of weed infestation reduction and potato yielding. A three-year field research study was conducted at the Agricultural Experimental Station Zawady (52°03′ N; 22°33′ E). The field experiment was set up in a split-plot design with three replicates. The first factor was the two edible potato cultivars, and the second was five methods of mechanical and chemical treatments with biostimulants: (1) control object—mechanical weeding, (2) the herbicide Avatar 293 ZC (clomazone + metribuzin), (3) the herbicide Avatar 293 ZC and the biostimulant PlonoStart, (4) the herbicide Avatar 293 ZC and the biostimulant Aminoplant, (5) the herbicide Avatar 293 ZC and the biostimulant Agro-Sorb Folium. Before emergence, the control object was double ridging and single ridging with harrowing and double ridging after emergence. On other objects (2–5) before emergence, two ridgings were applied. The assessment of weed infestation was performed using the quantitative weight method on two dates: about 2 weeks after the application of the herbicide and biostimulants and before the potato tubers harvest. The herbicide and the herbicide with biostimulants applied to the potato crop showed a positive influence in reducing the number and fresh weight of weeds compared to mechanical-only treatments. The highest effectiveness, calculated on the basis of the number and fresh mass of weeds, was found on the fifth object. It was, on average, 70.5 and 71.6% for cultivars, respectively. Research into the use of biostimulants in potato cultivation will continue due to the pro-environmental nature of biostimulants and the importance of this crop. It is certain that the methodology of future field experiments will meet the challenges of sustainable development.

Autonomous Navigation of a Forestry Robot Equipped with a Scanning Laser

This abstract is an overview of our research project entitled “Innovative Forest Plantation”, currently in progress. The aim of this project is to automate traditionally manual tasks for poplar plantations in the first years after planting, in particular mechanical weeding without the use of herbicides. The poplar forest is considered as a semi-structured environment where the dense canopy prevents the use of GPS signals and laser sensors are often preferred to localize the vehicle. In this paper, we focus on one of the main functionalities: autonomous navigation, which consists in detecting and locating trees to move safely in such complex environment. Autonomous navigation requires both a precise and robust mapping and localization solution. In this context, Simultaneous Localization and Mapping (SLAM) is very well-suited solution. The constructed map can be reliably used to plan semantic paths of the mobile robot in order treat specifically each tree. Simulations conducted on Gazebo and Robot Operation System (ROS) have proven that the robot could navigate autonomously in a poplar forest.

From ‘Miracle’ Rice Plants to Technology Hybridization: My SRI Journey

This paper is prepared based on the author’s experiences in working with SRI ideas and methods in diverse agroecological and socioeconomic contexts in Nepal and abroad. I have found that rice farmers used diverse field management strategies to incorporate SRI into their farming systems. Some farmers used all of the SRI practices introduced during their training, i.e., young seedlings, single seedlings, wider spacing, alternate wetting and drying irrigation, mechanical weeding, and the use of compost. However, the majority modified their methods to be appropriate for their farming situation. Farmers used younger seedlings in areas where irrigation was reliable and drainage facility was better. The use of mechanical weeding was very effective for higher yield; however, its effectiveness and productivity were not the same everywhere. Similarly, many farmers did not follow the advice to use compost (alone, or with fertilizer). It was interesting to note that the poorly-producing farmers were using more fertilizer than required. By contrast, the farmers who attended the SRI training have reduced their fertilizer use. In short, the introduction of SRI methods influenced the traditional rice farming system, but not in a uniform way. After years of experience, the majority of farmers adjusted these practices to fit their personal farming situation. Most farmers who changed their rice farming system were following neither SRI nor traditional practice, but rather a hybrid of methods, and they developed a hybrid system that is more feasible and productive in Nepal.

Adept to Adapt: Closing the Gender Capacity Gaps for Scaling Up System of Crop Intensification

The system of rice intensification (SRI) has been introduced as an alternative system for growing rice with lesser inputs and water. Labour is one of the most crucial concerns in the adoption of SRI by farmers. The adaptation to newer methods of rice cultivation is presumed to be easier if the workforce is adept to adapt. The new skills needed to adopt SRI, are transplanting young and single seedlings and use of mechanical weeders. The present study was therefore, undertaken, to identify training needs of women farmers in SRI cultivation. Line sowing was perceived as a new skill to be acquired and rated as highly skilful. Close attention was needed by women farmers/labour to systematically plant within the square. The paper attempts to apply the components of the community capitals framework (CCF) developed by Cornelia and Jan Flora (2013), for closing the gender capacity gaps in the uptake and scaling up of SCI/SRI. The seven forms of capital in this framework are considered not only as individual capabilities and endowments but are viewed as collective resources and are to be considered in the specific order of natural, cultural, human, social, political, financial and built capital. In this paper emphasis is being given to build the Human and Social Capital for closing the gender capacity gaps for scaling up SCI/SRI. There is immense scope of harnessing the potential of training members of women’s self-help groups (SHG) to form a SRI task force to help in the wide spread adoption of SRI by farmers.

Juxtaposition of Push-Type Wheel Hoe to Evaluate Its Weeding Coherence and Impair on Eggplant

Abstract: A study was done to access the best suited mechanical weed control method in eggplant crop as use of mechanical methods in eggplant crop will enhance the crop growth and reduce the use of herbicides and chemicals that affect the soil properties of crop. The weeding efficiency of the wheel hoe1 and wheel hoe2 was calculated at different levels of forward speed and depth of weeding in Allahabad region in eggplant crop at 21st, 40th and 75th day after sowing. The plant damage was recorded at the fifth day after the weeding operation was performed in the field. The plant damage increased as the speed of operation increased but the depth of operation had no significant effect on the plant damage. Plant damage was dependent upon the skill of the operator and environmental factors like sunshine, wind etc

Uji Efektivitas Herbisida Pasca Tumbuh Pada Tanaman Kedelai (Glycine max. L)

The purpose of this study was to determine the effect of weed control with herbicides on soybean growth and yield. This research was conducted in the Mojosari Agricultural Experimental Garden, Mojokerto, March - July 2021. The method used was a Randomized Block Design (RAK) with 3 treatments and 4 replications. The treatments were (P0) mechanical weeding (Control); (P1) Spraying herbicide 1X; (P2) spraying. 2X herbicide. selective herbicide spraying (Agil 100 ec) with the active ingredient propaquzafop 100 ec C-12 was carried out according to the treatment. Herbicide application had a better effect on soybean yields than control (mechanical weeding). However, spraying herbicides once had the same good effect as (P2) spraying 2X. The soybean plants tested were able to produce dry pods weight of 36.85 -39.00 g/plant, and dry weight 100 seeds 18,48-20,90 g.

Intelligent Weed Management Based on Object Detection Neural Networks in Tomato Crops

As the tomato (Solanum lycopersicum L.) is one of the most important crops worldwide, and the conventional approach for weed control compromises its potential productivity. Thus, the automatic detection of the most aggressive weed species is necessary to carry out selective control of them. Precision agriculture associated with computer vision is a powerful tool to deal with this issue. In recent years, advances in digital cameras and neural networks have led to novel approaches and technologies in PA. Convolutional neural networks (CNNs) have significantly improved the precision and accuracy of the process of weed detection. In order to apply on-the-spot herbicide spraying, robotic weeding, or precise mechanical weed control, it is necessary to identify crop plants and weeds. This work evaluates a novel method to automatically detect and classify, in one step, the most problematic weed species of tomato crops. The procedure is based on object detection neural networks called RetinaNet. Moreover, two current mainstream object detection models, namelyYOLOv7 and Faster-RCNN, as a one and two-step NN, respectively, were also assessed in comparison to RetinaNet. CNNs model were trained on RGB images monocotyledonous (Cyperus rotundus L., Echinochloa crus galli L., Setaria verticillata L.) and dicotyledonous (Portulaca oleracea L., Solanum nigrum L.) weeds. The prediction model was validated with images not used during the training under the mean average precision (mAP) metric. RetinaNet performed best with an AP ranging from 0.900 to 0.977, depending on the weed species. Faster-RCNN and YOLOv7 also achieved satisfactory results, in terms of mAP, particularly through data augmentation. In contrast to Faster CNN, YOLOv7 was less precise when discriminating monocot weed species. The results provide a better insight on how weed identification methods based on CNN can be made more broadly applicable for real-time applications.

Weeds Detection and Control in Rice Crop Using UAVs and Artificial Intelligence: A Review

Weeds are serious issues in rice farming and undesired plants that compete for water, light, space, and nutrients, reducing crop yields. Weeds competition in rice crops can result in yield failure of up to 100% if weeds are not controlled. Furthermore, weeds will raise protection costs by harbouring other pests such as diseases, insects, and nematodes that use weeds as alternate hosts. Rice fields are infested with grassy weeds, broadleaves, and sedges, among other weeds. Weeds detection is important to identify the types of weeds in rice areas and make the precise decision to determine the method of weeds control and reduce herbicide use. Chemical, biological, and mechanical weed control strategies such as manual weeding, mechanical weeding, and herbicide use are all part of the Integrated Weed Management (IWM) approach. In the rice field, however, pesticide spraying to eliminate the weed is a very common control approach. Still, this method has become ineffective due to frequently spraying with the same type of herbicide. Early detection is required to define the type of weeds in rice fields, make a precise decision on weed management methods, and prescribe the appropriate herbicide to the rice farmer. Artificial intelligence and unmanned aerial vehicles (UAVs) were largely applied to identify the weeds in rice fields and herbicide spraying to control the weeds. UAVs, such as drones, have recently shown to have a lot of potential in agriculture, such as crop health monitoring systems, assisting in planning irrigation schedules, estimating production data, and capturing weather analysis data and weed infestation. In Malaysia, UAVs are mostly utilized for nutrition and pesticide applications, particularly by smallholder farmers and industries. The integration use of artificial intelligence including unmanned aerial vehicles drones, and various sensors, hyperspectral, multispectral, and RGB (red-green-blue) cameras, thermal and odor sensor for weed early detection methods could ensure the possibility of a better outcome in managing weed problems. This paper reviews the detection and controlling of weeds using UAVs and artificial intelligence technologies in rice crop.