scholarly journals Automatic Pest Counting from Pheromone Trap Images Using Deep Learning Object Detectors for Matsucoccus thunbergianae Monitoring

Insects ◽  
2021 ◽  
Vol 12 (4) ◽  
pp. 342
Author(s):  
Suk-Ju Hong ◽  
Il Nam ◽  
Sang-Yeon Kim ◽  
Eungchan Kim ◽  
Chang-Hyup Lee ◽  
...  

The black pine bast scale, M. thunbergianae, is a major insect pest of black pine and causes serious environmental and economic losses in forests. Therefore, it is essential to monitor the occurrence and population of M. thunbergianae, and a monitoring method using a pheromone trap is commonly employed. Because the counting of insects performed by humans in these pheromone traps is labor intensive and time consuming, this study proposes automated deep learning counting algorithms using pheromone trap images. The pheromone traps collected in the field were photographed in the laboratory, and the images were used for training, validation, and testing of the detection models. In addition, the image cropping method was applied for the successful detection of small objects in the image, considering the small size of M. thunbergianae in trap images. The detection and counting performance were evaluated and compared for a total of 16 models under eight model conditions and two cropping conditions, and a counting accuracy of 95% or more was shown in most models. This result shows that the artificial intelligence-based pest counting method proposed in this study is suitable for constant and accurate monitoring of insect pests.

2021 ◽  
Vol 25 (1) ◽  
pp. 1-22
Author(s):  
MP Ali ◽  
B Nessa ◽  
MT Khatun ◽  
MU Salam ◽  
MS Kabir

The damage caused by insect pest is the continual factor for the reduction of rice production. To date, 232 rice insect pest species are identified in Bangladesh and more than 100 species of insects are considered pests in rice production systems globally, but only about 20 - 33 species can cause significant economic loss. The major goal of this study is to explore all the possible ways of developed and proposed technologies for rice insect pests management and minimize economic losses. Insect pests cause 20% average yield loss in Asia where more than 90% of the world's rice is produced. In Bangladesh, outbreak of several insects such as rice hispa, leafroller, gallmidge, stem borers and brown planthopper (BPH) occurs as severe forms. Based on previous reports, yield loss can reach upto 62% in an outbreak situation due to hispa infestation. However, BPH can cause 44% yield loss in severe infestested field. To overcome the outbreaks in odd years and to keep the loss upto 5%, it is necessary to take some preventive measures such as planting of resistant or tolerant variety, stop insecticide spraying at early establishment of rice, establish early warning and forecasting system, avoid cultivation of susceptible variety and following crop rotation. Subsequent quick management options such as insecticidal treatment for specific insect pest should also be broadcasted through variety of information systems. Advanced genomic tool can be used to develop genetically modified insect and plants for sustainable pest management. In addition, to stipulate farmers not use insecticides at early crop stgae and minimize general annualized loss, some interventions including training rice farmers, regular field monitoring, digitalization in correct insect pests identification and their management (example; BRRI rice doctor mobile app), and demonstration in farmers field. Each technology itself solely or combination of two or more or all the packages can combat the insect pests, save natural enemies, harvest expected yield and contribute to safe food production in Bangladesh. Bangladesh Rice J. 25 (1) : 1-22, 2021


2019 ◽  
Vol 2 (1) ◽  
pp. 238-243
Author(s):  
Anjali Gyawali ◽  
Bandana Regmi ◽  
Rameshwor Pudasaini ◽  
Namuna Acharya

A study on diversity and abundance of insects in rice field was conducted at farmer field of Lamahi, Dang during July to October in 2019. Insects were collected using sweep net and light trap. Overall, 414 insect specimen representing 11 families and 8 orders were collected during the period. Grasshopper (23.98%) with including all species was the most abundance insect found in rice field as it followed by brown plant hopper (16.62%). Among the eight insect orders captured Orthoptera (29.16%) was the most abundance insect order followed by Homoptera (16.62%). As the diversity of insect pest in this area may responsible economic losses was found which will be useful to adapt appropriate management practices to keep them at normal area. The presence of natural enemies should conserve to enhance the natural biological control of insect pests.


2017 ◽  
Vol 8 (1) ◽  
Author(s):  
Clement Akotsen-Mensah ◽  
Isaac N. Ativor ◽  
Roger S. Anderson ◽  
Kwame Afreh-Nuamah ◽  
Collison F. Brentu ◽  
...  

Abstract Mango farmers in Ghana are confronted with many pest problems like fruit flies, Sternochetus mangiferae (F.), and mealy bugs. Different pest management options are available to mango farmers; however, the extent to which they apply the available pest management options is not well known. A survey was conducted among 60 farmers in southeastern Ghana, from October–December 2015 mango season, to find out the level of knowledge and practice of insect pest management used by mango farmers. The results showed that most farmers use conventional insecticides to control insect pests in mango. Majority of the farmers (30%) use a composite insecticide (Cydim super; 36 g cypermethrin + 400 g dimethoate per liter), whereas 3.3% use Pyrinex (chlorpyrifos 480 g/liter). Majority of insecticides used belong to WHO category II. Ninety percent (90%) of the farmers use cultural practices and pheromone traps. Pheromone traps are, however, used for fruit flies but not for S. mangiferae. Over 80% of the respondents who used pesticides to control pests have also adopted GLOBALGAP standards for certification. The results are discussed based on the importance of adoption of IPM strategies in mango production and the possible reduction of fruit rejection during mango export in Ghana.


2020 ◽  
pp. 1-10
Author(s):  
Nana Millicent Duduzile Buthelezi ◽  
Tieho Paulus Mafeo ◽  
Nhlanhla Mathaba

Preharvest factors such as poor orchard management and field sanitation can lead to pathological infection of the tree fruit being grown as well as insect pest infestation, resulting in poor postharvest fruit quality. Wind and hail damage may cause significant tree fruit abrasions and blemishes. Consequently, these preharvest factors may reduce yield and cause market and economic losses. One of the most successful methods used to manage tree fruit pathogens and insect infestation is the application of agrochemicals, predominantly fungicides and insecticides. However, this method has recently been criticized due to the adverse effects on field workers’ safety, consumers’ health, and the environment. The development and use of preharvest bagging are among the most environmentally friendly technologies intended for safe enhancement of tree fruit quality. The technique protects tree fruit against pathogens, insect pests, physiological disorders, agrochemical residues, fruit abrasions, sunburn, and bird damage, and it further modifies the microenvironment for fruit development with its various beneficial effects on its external and internal quality. Furthermore, because of the global restrictions of agrochemicals and social awareness, this technique provides extensive relief to growers and consumers. However, bagging is labor-intensive and expensive; therefore, its benefits or advantages and disadvantages must be thoroughly investigated if it is to be promoted commercially. This review examines the improvement of tree fruit quality by the application of preharvest bagging during early stages of fruit growth and development. The latest advances in the development and use of tree fruit bagging and its economic impact and cost–benefit ratio are discussed, as are recommendations for the formulation of bagging materials that could be valuable in the future.


2021 ◽  
Vol 19 (1) ◽  
Author(s):  
Vaijayanti A. Tamhane ◽  
Surhud S. Sant ◽  
Abhilash R. Jadhav ◽  
Abdul R. War ◽  
Hari C. Sharma ◽  
...  

Abstract Background Spotted stem borer- Chilo partellus - a Lepidopteran insect pest of Sorghum bicolor is responsible for major economic losses. It is an oligophagous pest, which bores through the plant stem, causing ‘deadheart’ and hampering the development of the main cob. We applied a label-free quantitative proteomics approach on three genotypes of S. bicolor with differential resistance/ susceptibility to insect pests, intending to identify the S. bicolor’s systemic protein complement contributing to C. partellus tolerance. Methods The proteomes of S. bicolor with variable resistance to insect pests, ICSV700, IS2205 (resistant) and Swarna (susceptible) were investigated and compared using label-free quantitative proteomics to identify putative leaf proteins contributing to resistance to C. partellus. Results The multivariate analysis on a total of 967 proteins led to the identification of proteins correlating with insect resistance/susceptibility of S. bicolor. Upon C. partellus infestation S. bicolor responded by suppression of protein and amino acid biosynthesis, and induction of proteins involved in maintaining photosynthesis and responding to stresses. The gene ontology analysis revealed that C. partellus-responsive proteins in resistant S. bicolor genotypes were mainly involved in stress and defense, small molecule biosynthesis, amino acid metabolism, catalytic and translation regulation activities. At steady-state, the resistant S. bicolor genotypes displayed at least two-fold higher numbers of unique proteins than the susceptible genotype Swarna, mostly involved in catalytic activities. Gene expression analysis of selected candidates was performed on S. bicolor by artificial induction to mimic C. partellus infestation. Conclusion The collection of identified proteins differentially expressed in resistant S. bicolor, are interesting candidates for further elucidation of their role in defense against insect pests.


Author(s):  
S. Khan ◽  
P. K. Gupta

<p><strong>Abstract.</strong> Tree counting can be a challenging and time consuming task, especially if done manually. This study proposes and compares three different approaches for automatic detection and counting of trees in different vegetative regions. First approach is to mark extended minima’s, extended maxima’s along with morphological reconstruction operations on an image for delineation and tree crown segmentation. To separate two touching crowns, a marker controlled watershed algorithm is used. For second approach, the color segmentation method for tree identification is used. Starting with the conversion of an RGB image to HSV color space then filtering, enhancing and thresholding to isolate trees from non-trees elements followed by watershed algorithm to separate touching tree crowns. Third approach involves deep learning method for classification of tree and non-tree, using approximately 2268 positive and 1172 negative samples each. Each segment of an image is then classified and sliding window algorithm is used to locate each tree crown. Experimentation shows that the first approach is well suited for classification of trees is dense vegetation, whereas the second approach is more suitable for detecting trees in sparse vegetation. Deep learning classification accuracy lies in between these two approaches and gave an accuracy of 92% on validation data. The study shows that deep learning can be used as a quick and effective tool to ascertain the count of trees from airborne optical imagery.</p>


2022 ◽  
Vol 12 (1) ◽  
Author(s):  
Wonsub Yun ◽  
J. Praveen Kumar ◽  
Sangjoon Lee ◽  
Dong-Soo Kim ◽  
Byoung-Kwan Cho

AbstractThe prevention of the loss of agricultural resources caused by pests is an important issue. Advances are being made in technologies, but current farm management methods and equipment have not yet met the level required for precise pest control, and most rely on manual management by professional workers. Hence, a pest detection system based on deep learning was developed for the automatic pest density measurement. In the proposed system, an image capture device for pheromone traps was developed to solve nonuniform shooting distance and the reflection of the outer vinyl of the trap while capturing the images. Since the black pine bast scale pest is small, pheromone traps are captured as several subimages and they are used for training the deep learning model. Finally, they are integrated by an image stitching algorithm to form an entire trap image. These processes are managed with the developed smartphone application. The deep learning model detects the pests in the image. The experimental results indicate that the model achieves an F1 score of 0.90 and mAP of 94.7% and suggest that a deep learning model based on object detection can be used for quick and automatic detection of pests attracted to pheromone traps.


2020 ◽  
Vol 26 (1) ◽  
Author(s):  
Priya Lokare ◽  
Sumia Fatima

Mango saplings go through the many insect pests, fungal, bacterial diseases during nursery condition and these symptoms will persist till flowering and fruiting period and result in the huge economic losses. Majority mango saplings couldn’t reach upto flowering and fruiting stage it dies in the nursery conditions. This is major threat to the nursery owners because mango saplings having great demand all over the year, therefore buyers refuse to purchase diseased saplings. In the recent years the disease becomes severe in nursery plants, on young leaves, symptoms appear as irregular black necrotic spots on both sides. Pathogen present on the infected leaves, twig and fallen leaves serves as the major source of infection and spreads by rain splashed conidia. Survey was carried out to know the prevalence of diseases in nursery conditions for that Sanket Nursery Wakadi, Taluka Rahta was selected. There were 4 varieties of mango found in Sanket Nursery that were, Keshar, Payari, Mallika and Ratna. During the survey various fungal and insect pest diseases were observed. Anthracnose symptoms caused by Colletotrichum gloeosporioides, little leaf notcher, coconut scale, mango gall midge, white mango scale, stem blight, powdery mildew, hairy caterpillar etc. were found in large scale.


2020 ◽  
Vol 31 (1) ◽  
pp. 24-35 ◽  
Author(s):  
Somiahnadar Rajendran

Insects are a common problem in stored produce. The author describes the extent of the problem and approaches to countering it. Stored products of agricultural and animal origin, whether edible or non-edible, are favourite food for insect pests. Durable agricultural produce comprising dry raw and processed commodities and perishables (fresh produce) are vulnerable to insect pests at various stages from production till end-use. Similarly, different animal products and museum objects are infested mainly by dermestids. Insect pests proliferate due to favourable storage conditions, temperature and humidity and availability of food in abundance. In addition to their presence in food commodities, insects occur in storages (warehouses, silos) and processing facilities (flour mills, feed mills). Insect infestation is also a serious issue in processed products and packed commodities. The extent of loss in stored products due to insects varies between countries depending on favourable climatic conditions, and pest control measures adopted. In stored food commodities, insect infestation causes loss in quantity, changes in nutritional quality, altered chemical composition, off-odours, changes in end-use products, dissemination of toxigenic microorganisms and associated health implications. The insects contribute to contaminants such as silk threads, body fragments, hastisetae, excreta and chemical secretions. Insect activity in stored products increases the moisture content favouring the growth of moulds that produce mycotoxins (e.g., aflatoxin in stored peanuts). Hide beetle, Dermestes maculatus infesting silkworm cocoons has been reported to act as a carrier of microsporidian parasite Nosema bombycis that causes pebrine disease in silkworms. In dried fish, insect infestation leads to higher bacterial count and uric acid levels. Insects cause damage in hides and skins affecting their subsequent use for making leather products. The trend in stored product insect pest management is skewing in favour of pest prevention, monitoring, housekeeping and finally control. Hermetic storage system can be supplemented with CO2 or phosphine application to achieve quicker results. Pest detection and monitoring has gained significance as an important tool in insect pest management. Pheromone traps originally intended for detection of infestations have been advanced as a mating disruption device ensuing pest suppression in storage premises and processing facilities; pheromones also have to undergo registration protocols similar to conventional insecticides in some countries. Control measures involve reduced chemical pesticide use and more non-chemical inputs such as heat, cold/freezing and desiccants. Furthermore, there is an expanding organic market where physical and biological agents play a key role. The management options for insect control depend on the necessity or severity of pest incidence. Generally, nonchemical treatments, except heat, require more treatment time or investment in expensive equipment or fail to achieve 100% insect mortality. Despite insect resistance, environmental issues and residue problems, chemical control is inevitable and continues to be the most effective and rapid control method. There are limited options with respect to alternative fumigants and the alternatives have constraints as regards environmental and health concerns, cost, and other logistics. For fumigation of fresh agricultural produce, new formulations of ethyl formate and phosphine are commercially applied replacing methyl bromide. Resistance management is now another component of stored product pest management. In recent times, fumigation techniques have improved taking into consideration possible insect resistance. Insect control deploying nanoparticles, alone or as carriers for other control agents, is an emerging area with promising results. As there is no single compound with all the desired qualities, a necessity has arisen to adopt multiple approaches. Cocktail applications or combination treatments (IGRs plus organophosphorus insecticides, diatomaceous earth plus contact insecticides, nanoparticles plus insecticides/pathogens/phytocompounds and conventional fumigants plus CO2; vacuum plus fumigant) have been proved to be more effective. The future of store product insect pest management is deployment of multiple approaches and/or combination treatments to achieve the goal quickly and effectively.


Electronics ◽  
2021 ◽  
Vol 10 (10) ◽  
pp. 1161
Author(s):  
Kuo-Hao Fanchiang ◽  
Yen-Chih Huang ◽  
Cheng-Chien Kuo

The safety of electric power networks depends on the health of the transformer. However, once a variety of transformer failure occurs, it will not only reduce the reliability of the power system but also cause major accidents and huge economic losses. Until now, many diagnosis methods have been proposed to monitor the operation of the transformer. Most of these methods cannot be detected and diagnosed online and are prone to noise interference and high maintenance cost that will cause obstacles to the real-time monitoring system of the transformer. This paper presents a full-time online fault monitoring system for cast-resin transformer and proposes an overheating fault diagnosis method based on infrared thermography (IRT) images. First, the normal and fault IRT images of the cast-resin transformer are collected by the proposed thermal camera monitoring system. Next is the model training for the Wasserstein Autoencoder Reconstruction (WAR) model and the Differential Image Classification (DIC) model. The differential image can be acquired by the calculation of pixel-wise absolute difference between real images and regenerated images. Finally, in the test phase, the well-trained WAR and DIC models are connected in series to form a module for fault diagnosis. Compared with the existing deep learning algorithms, the experimental results demonstrate the great advantages of the proposed model, which can obtain the comprehensive performance with lightweight, small storage size, rapid inference time and adequate diagnostic accuracy.


Sign in / Sign up

Export Citation Format

Share Document