Toward Sustainability: Trade-Off Between Data Quality and Quantity in Crop Pest Recognition

The crop pest recognition based on the convolutional neural networks is meaningful and important for the development of intelligent plant protection. However, the current main implementation method is deep learning, which relies heavily on large amounts of data. As known, current big data-driven deep learning is a non-sustainable learning mode with the high cost of data collection, high cost of high-end hardware, and high consumption of power resources. Thus, toward sustainability, we should seriously consider the trade-off between data quality and quantity. In this study, we proposed an embedding range judgment (ERJ) method in the feature space and carried out many comparative experiments. The results showed that, in some recognition tasks, the selected good data with less quantity can reach the same performance with all training data. Furthermore, the limited good data can beat a lot of bad data, and their contrasts are remarkable. Overall, this study lays a foundation for data information analysis in smart agriculture, inspires the subsequent works in the related areas of pattern recognition, and calls for the community to pay more attention to the essential issue of data quality and quantity.

RNAi Targets in Agricultural Pest Insects: Advancements, Knowledge Gaps, and IPM

The unprecedented target-specificity of double-stranded RNA (dsRNA), due to its sequence-specific mode of action, puts dsRNA at the forefront of biosafe insecticide technology. Since 2007, sensitive target genes have been identified in numerous crop pest insects, with an end goal of applying RNA interference (RNAi) in pest management. Key RNAi targets identified include genes involved in (1) feeding and digestion, (2) production of dsRNases, (3) resistance to insecticides and plant allelochemicals, (4) reproductive fitness, and (5) transmission of plant viruses. Despite the advances, there remain critical knowledge gaps in each of these areas. Particular emphasis must be placed on ensuring RNAi's compatibility with integrated pest management (IPM), via further identification of molecular targets that reduce crop damage while sustaining pest (host) populations for highly specialized biocontrol agents, the latter representing a core pillar of IPM.

Identification and Area Calculation of the Egg Mass of Spodoptera Frugiperda Based on Faster-RCNN

Spodoptera frugiperda is a migratory and destructive crop pest. The number of eggs is an important method to evaluate the pest situation, which can be estimated by the area of egg mass. The traditional manual method is inefficient, but the new method of egg mass image recognition improves the efficiency of eggs number estimation. In this paper, the optimized Faster-RCNN target detection algorithm was used to recognize the egg mass image of Spodoptera frugiperda. The Maximum Between-Class Variance method (Otsu) was used for threshold segmentation to obtain the position, shape and size of the egg mass and calculate the area of the egg mass. The mean value of the relative error of the egg mass area in the test samples was -0.02032, and the minimum value was -0.00047. The experimental results show that the egg area calculation method proposed in this paper is fast and accurate, which can meet the requirements of egg area measurement.

Selected Diptera of City Park Kolmanka, Prešov (Slovakia)

In total 65 Diptera species from 20 families (Anisopodidae (2 spp.), Asilidae (1), Bibionidae (1), Clusiidae (1), Culicidae (8), Dolichopodidae (7), Drosophilidae (4), Dryomyzidae (1), Empididae (2), Heleomyzidae (5), Hybotidae (5), Lauxaniidae (4), Limoniidae (9), Opomyzidae (2), Pallopteridae (2), Psychodidae (6), Rhagionidae (2), Scatopsidae (1), Trichoceridae (1) and Ulidiidae (1)) were recorded. The species Drapetis flavipes Macquart, 1834 (Hybotidae), is recorded for the first time in Slovakia. Ten species belong among uncommon or rare (namely: Atypophthalmus (Atypophthalmus) inustus (Meigen, 1818), Calliopum splendidum Papp, 1978, Dioctria linearis (Fabricius, 1787), Cheilotrichia (Empeda) neglecta (Lackschewitz, 1927), Chrysopilus asiliformis (Preyssler, 1791), Ochlerotatus (Ochlerotatus) nigrinus (Eckstein 1918), Philosepedon (Philosepedon) austriacum Vaillant, 1974, Suillia variegata (Loew, 1862), Toxoneura modesta (Meigen, 1830) and Trichomyia urbica Curtis, 1839). On the other hand, two invasive species are also reported. Drosophila (Sophophora) suzukii (Matsumura, 1931) is an invasive crop pest and Aedes (Finlaya) japonicus japonicus (Theobald, 1901) is an invasive biting pest, a potential vector for various diseases. City parks are also important from the point of view of Diptera biodiversity and more attention needs to be paid to them.

A Model for Pest Control using Integrated Approach: Impact of Latent and Gestation Delays

In this article, we have established a mathematical model using impulsive differential equations for the dynamics of crop pest management in the presence of a pest with its predator and bio-pesticides. The pest population is divided into two subpopulations, namely, the susceptible pests and the infected pests. In this control process, bio-pesticides (generally virus) infect the susceptible pest through viral infection within the pest and make it infected so that predators can consume it quickly. We assume that pest controlling, using this integrated approach, is a delayed process and thus incorporated latent time of susceptible pest and gestation delay of predator in the model as time delay parameters. The system dynamics have been analyzed using qualitative theory: the existence of the equilibrium points and their stability properties has been derived. Hopf bifurcation of the coexisting equilibrium point is presented for both the delayed and non-delayed system. Detail numerical simulations are performed in support of analytical results and illustrate the different dynamical regimes observed in the system. We have observed that the system becomes free of infection when the latent time of the pest is large. Coexisting equilibrium exists for the lower value of latent delay, and it can change the stability properties from stable to unstable when it crosses its critical value. In contrast, gestation delay affects the stability switches of coexisting equilibrium only. The combined effect of the two delays on the system is shown numerically. Also, viral replication rate, infection rate (from virus to pest) is also significant from the pest management perspective. In summary, both the delay is essential for crop pest management, and pest control will be successful with tolerable delays.

IFSG: Intelligence agriculture crop-pest detection system using IoT automation system

The agricultural and technological combination is blessed for modern world life. Internet of things (IoT) is essential for comfort and development to our agriculture side. In our study, we detected the various pest using different types of sensors and this information has automatically sent to the farmer's mobile for the alert. All these sensors had a central database. Those sensors collect all the data and display the results compared to the central data. The High-image sensor will be able to detect all the rays emitted from the plant and another one is the gas sensor which is able to detect all the gases coming from the diseased plant. We mainly use sound sensor, MQ138, CMOSOV-7670, AMG-8833 for a better automation system. We test it with real-time environment conditions (40°C≤TA≤14°C). Crop pest detection automatic process is more efficient than the other detection process according to testing output. As a result, far-reaching changes in the agricultural sector are possible. To reduce extra cost and increasing more farming ability we need to IoT and Agriculture combinations more.

GIS based pest-weather model to predict the incidence of Girdle beetle (Oberiopsis brevis) in Soybean crop

Girdle beetle (Oberiopsis brevis) is an important insect of soybean that can cause up to 42.2% yield loss in severe infestation during flowering stage. The infestation of girdle beetle is prevailed by congenial environmental conditions, which leads girdle beetle to be the severe pest of soybean. The present study assesses the relevant weather variables that can cause the peak infestation. Crop Pest Surveillance and Advisory Project (CROPSAP) survey data of girdle beetle incidence were analyzed with weather variables using correlation and regression techniques. The girdle beetle infestation had significantly positive correlation with relative humidity of current and 2nd lag week (RH0, RH-2); and with rainfall of 2nd lag week (RF-2) but significantly negative correlation with maximum temperature of 1st lag week (TMax-1). The multiple regression technique was used to develop the forewarning models for three zones (Vidarbha, Madhya Maharashtra and Marathwada zones) and overall Maharashtra, the developed models could explain 80.30%, 94.62%, 73.56% and 79.56% variation in girdle beetle infestation, respectively. The congenial conditions for the peak infestation of girdle beetle on soybean have been worked out and validated, which were TMax0, RH0, RF0, RH-1, RF-1, TMax-2, and RF-2 ranged between 28.6-31.6 ºC, 85.2- 91.8 %, 31.8-119.2 mm, 86.3-92.6 %, 38.1-76.4 mm, 27.7-30.8ºC, and 23.3-60.7 mm, respectively. The insect forewarning would be useful in devising the integrated management strategies for protecting the crop from insect in the incidence region.

Research Progress of Crop Disease Image Recognition Based on Wireless Network Communication and Deep Learning

The traditional digital image processing technology has its limitations. It requires manual design features, which consumes manpower and material resources, and identifies crops with a single type, and the results are bad. Therefore, to find an efficient and fast real-time disease image recognition method is very meaningful. Deep learning is a machine learning algorithm that can automatically learn representative features to achieve better results in areas of image recognition. Therefore, the purpose of this paper is to use deep learning methods to identify crop pests and diseases and to find efficient and fast real-time image recognition methods of disease. Deep learning is a newly developed discipline in recent years. Its purpose is to study how to actively obtain a variety of feature representation methods from data samples and rely on data-driven methods, a series of nonlinear transformations are applied to finally collect the original data from specific to abstract, from general to specified semantics, and from low-level to high-level characteristic forms. This paper analyzes the classical and the latest neural network structure based on the theory of deep learning. For the problem that the network based on natural image classification is not suitable for crop pest and disease identification tasks, this paper has improved the network structure that can take care of both recognition speed and recognition accuracy. We discussed the influence of the crop pest and disease feature extraction layer on recognition performance. Finally, we used the inner layer as the main structure to be the pest and disease feature extraction layer by comparing the advantages and disadvantages of the inner and global average pooling layers. We analyze various loss functions such as Softmax Loss, Center Loss, and Angular Softmax Loss for pest identification. In view of the shortcomings of difficulty in loss function training, convergence, and operation, making the distance between pests and diseases smaller and the distance between classes more greater improved the loss function and introduced techniques such as feature normalization and weight normalization. The experimental results show that the method can effectively enhance the characteristic expression ability of pests and diseases and thus improve the recognition rate of pests and diseases. Moreover, the method makes the pest identification network training simpler and can improve the pest and disease recognition rate better.