Pepper leaf disease recognition based on enhanced lightweight convolutional neural networks.

Pepper leaf disease identification based on convolutional neural networks (CNNs) is one of the interesting research areas. However, most existing CNN-based pepper leaf disease detection models are suboptimal in terms of accuracy and computing performance. In particular, it is challenging to apply CNNs on embedded portable devices due to a large amount of computation and memory consumption for leaf disease recognition in large fields. Therefore, this paper introduces an enhanced lightweight model based on GoogLeNet architecture. The initial step involves compressing the Inception structure to reduce model parameters, leading to a remarkable enhancement in recognition speed. Furthermore, the network incorporates the spatial pyramid pooling structure to seamlessly integrate local and global features. Subsequently, the proposed improved model has been trained on the real dataset of 9183 images, containing 6 types of pepper diseases. The cross-validation results show that the model accuracy is 97.87%, which is 6% higher than that of GoogLeNet based on Inception-V1 and Inception-V3. The memory requirement of the model is only 10.3 MB, which is reduced by 52.31%-86.69%, comparing to GoogLeNet. We have also compared the model with the existing CNN-based models including AlexNet, ResNet-50 and MobileNet-V2. The result shows that the average inference time of the proposed model decreases by 61.49%, 41.78% and 23.81%, respectively. The results show that the proposed enhanced model can significantly improve performance in terms of accuracy and computing efficiency, which has potential to improve productivity in the pepper farming industry.

A New Pest Detection Method Based on Improved YOLOv5m.

Pest detection in plants is essential for ensuring high productivity. Convolutional neural networks (CNN)-based deep learning advancements recently have made it possible for researchers to increase object detection accuracy. In this study, pest detection in plants with higher accuracy is proposed by an improved YOLOv5m-based method. First, the SWin Transformer (SWinTR) and Transformer (C3TR) mechanisms are introduced into the YOLOv5m network so that they can capture more global features and can increase the receptive field. Then, in the backbone, ResSPP is considered to make the network extract more features. Furthermore, the global features of the feature map are extracted in the feature fusion phase and forwarded to the detection phase via a modification of the three output necks C3 into SWinTR. Finally, WConcat is added to the fusion feature, which increases the feature fusion capability of the network. Experimental results demonstrate that the improved YOLOv5m achieved 95.7% precision rate, 93.1% recall rate, 94.38% F1 score, and 96.4% Mean Average Precision (mAP). Meanwhile, the proposed model is significantly better than the original YOLOv3, YOLOv4, and YOLOv5m models. The improved YOLOv5m model shows greater robustness and effectiveness in detecting pests, and it could more precisely detect different pests from the dataset.

Appraisal of stormwater-induced runoff quality influenced by site-specific land use patterns in the south-eastern region of Bangladesh.

Urban stormwater runoff is considered as one of the major contributors to nonpoint source that contributes to the pollution of all water resources in the surrounding environment. Pollutant concentrations of urban stormwater runoff are directly or indirectly linked with land use types in a catchment that is quite different in different places, and hence, site-specific studies are necessary, unless otherwise the modelling of runoff quality using modelling tools may not be rationally reflected the actual field scenarios. This paper portrays the influence of land use types on stormwater runoff physicochemical quality in Chattogram city, where land use's demarcation is often complicated due to the different natural and human-induced anthropogenic factors. The stormwater runoff samples were collected from the residential, commercial, institutional, and industrial land use types, in the city of Chattogram, Bangladesh, during the period from July to September 2020. The standard laboratory protocol for elemental concentrations and principal component analysis was performed in addition to basic statistics to identify the influence of urban surface characteristics on the quality of stormwater runoff. In general, pollutant concentrations were identified by analysing key physical and chemical quality parameters including colour, temperature, turbidity, total suspended solids (TSS), total dissolved solids (TDS), electrical conductivity (EC), salinity, pH, dissolved oxygen (DO), biochemical oxygen demand (BOD5), chemical oxygen demand (COD) and ammonia nitrogen (NH3-N) and exhibited 2 to 3 times higher concentrations than reported elsewhere. Furthermore, the present study reported the greater concentrations of few pollutants, such as TSS, BOD5, and EC, derived from the residential land uses compared to other land use types that are surprising; however, it confirmed the distinct complexity of unplanned land use patterns that should not be overlooked rather discussed. The strong correlation between land use types and stormwater runoff quality indicates the site-specific influences of runoff quality. The outcomes of this study would be particularly helpful in calibrating water quality models considering different land use types. Additionally, datasets and information obtained from this research will assist engineers and practitioners in developing decision-making tools for effective stormwater management.

Phodopus roborovskii SH101 as a systemic infection model of SARS-CoV-2

Severe acute respiratory syndrome CoV-2 (SARS-CoV-2) is currently causing a worldwide threat with its unusually high transmission rates and rapid evolution into diverse strains. Unlike typical respiratory viruses, SARS-CoV-2 frequently causes systemic infection by breaking the boundaries of the respiratory systems. The development of animal models recapitulating the clinical manifestations of COVID-19 is of utmost importance not only for the development of vaccines and antivirals but also for understanding the pathogenesis. However, there has not been developed an animal model for systemic infection of SARS-CoV-2 representing most aspects of the clinical manifestations of COVID-19 with systemic symptoms. Here we report that a hamster strain of Phodopus roborovskii SH101, a laboratory inbred hamster strain of P. roborovskii, displayed most symptoms of systemic infection upon SARS-CoV-2 infection as in the case of the human counterpart, unlike current COVID-19 animal models. P. roborovskii SH101 post-infection of SARS-CoV-2 represented most clinical symptoms of COVID-19 such as snuffling, dyspnea, cough, labored breathing, hunched posture, progressive weight loss, and ruffled fur, in addition to high fever following shaking chills. Histological examinations also revealed a serious right-predominated pneumonia as well as slight organ damages in the brain and liver, manifesting systemic COVID-19 cases. Considering the merit of a small animal as well as its clinical manifestations of SARS-CoV-2 infection in human, this hamster model seems to provide an ideal tool to investigate COVID-19. Author summaryAlthough the current animal models supported SARS-CoV-2 replication and displayed varying degrees of illness after SARS-CoV-2 infection, the infections of SARS-CoV-2 were mainly limited to the respiratory systems of these animals, including hACE2 transgenic mice, hamsters, ferrets, fruit bats, guinea pigs, African green monkey, Rhesus macaques, and Cynomolgus macaques. While these animal models can be a modest model for the respiratory infection, there is a clear limit for use them in the study of COVID-19 that also displays multiple systemic symptoms. Therefore, the development of an animal model recapitulating COVID-19-specific symptoms such as the right-predominated pneumonia would be the utmost need to overcome the imminent threat posed by COVID-19. We identified a very interesting hamster strain, Phodopus roborovskii SH101, which mimics almost all aspects of the clinical manifestations of COVID-19 upon SARS-CoV-2 infection. Unlike the current animal models, SARS-CoV-2-infected P. roborovskii SH101 not only displayed the symptoms of respiratory infection but also clinical manifestations specific to human COVID-19 such as high fever following shaking chills, serious right-predominated pneumonia, and minor organ damages in the brain and liver.

Optimal damping for generalized unified power flow controller equipped single machine infinite bus system for addressing low frequency oscillation.

Low frequency oscillation (LFO) is one of the major concerns for reliable operation of the power system. This LFO occurs due to the failure of the rotor to supply sufficient damping torque to compensate the imbalance between mechanical input and electrical output. Hence, in this paper, we adopt a third generation flexible AC transmission system (FACTS) device named generalized unified power flow controller (GUPFC) based damping controller in order to investigate its effect for mitigating LFO for an single machine infinite bus (SMIB) system. To find an effective damping controller-optimizer pair, we integrate proportional-integral (PI) or lead-lag as a controller and grey wolf optimizer (GWO), differential evolution (DE), particle swarm optimization (PSO), whale optimization algorithm (WOA), and chaotic whale optimization algorithm (CWOA) as an optimizer. Later, we investigate the performances for the above mentioned controller-optimizer pairs through time domain simulation, eigenvalue analysis, nyquist stability test, and quantitative analysis. Moreover, we carry out two non-parametric statistical tests named as one sample Kolmogorov-Smirnov (KS) test and paired sample t-test to identify statistical distribution as well as uniqueness of our optimization algorithms. Our analyses reveal that the GWO tuned lead-lag controller surpasses all other controller-optimizer combinations.

Spatiotemporal trends of selected air quality parameters during force lockdown and its relationship to COVID-19 positive cases in Bangladesh.

Worldwide improved air quality in different cities is reported influenced by lockdown came in force due to COVID-19 pandemic; however, as expected, such changes might have been different in different places. And what is still not very clear whether air quality pollutants have some link to account COVID-19 positive cases and death tolls. This study aims to evaluate the spatiotemporal variability of air pollutants and their relationship to COVID-19 positive cases in major cities in Bangladesh. The relevant data of air pollutants and COVID-19 positive cases are collected, analyzed, discussed for lockdown period of 26 March to 26 April 2020 in comparison to data for same period averaging over 2013 to 2019 for eight major cities in Bangladesh. To characterize air pollutants affected by lockdown, trend and rate of changes were carried out using Mann-Kandle and Sen's slope methods, while spatial variability across the cities was done using ArcGIS and statistics within ArcGIS. The substantial reduction of mean concentrations in the range of 30-65%, 20-80%, 30 - 80%, 65 - 90% and 75 - 90% across the cities is found during lockdown compared to typical mean in previous years for the PM2.5, PM10, SO2, CO, and NO2 concentrations in air. Among the cities studied, it is seen that relatively lesser reduction in Dhaka, Gazipur and Narayanganj and moderate reduction in Chittagong, Rajshahi, Khulna and Barisal, while significantly bigger reduction in Sylhet influenced by the city attributes and climatic variabilities. Among all the pollutants studied, the increasing trends of NO2 and CO in Dhaka, Gazipur and Narayanganj are unexpected even in lockdown pointing the effectiveness of lockdown management. Correlation among the air pollutants and confirmed COVID-19 cases across the cities depict foggy relationship, while PCA integrated over the cities revealed association with gaseous pollutants pointing stronger effects of NO2. This relationship illustrates air pollution health effects may increase vulnerability to COVID-19 cases.

Emblica officinalis improves glycemic status and oxidative stress in STZ induced type 2 diabetic model rats

OBJECTIVE@#To evaluate the antidiabetic and antioxidant potential of Emblica officinalis (E. officinalis) fruit on normal and type 2 diabetic rats.@*METHODS@#Type 2 diabetes was induced into the male Long-Evans rats. The rats were divided into nine groups including control groups receiving water, type 2 diabetic controls, type 2 diabetic rats treated with glibenclamide (T2GT) and type 2 diabetic rats treated with aqueous extract of fruit pulp of E. officinalis. They were fed orally for 8 weeks with a single feeding. Blood was collected by cutting the tail tip on 0 and 28 days and by decapitation on 56 day. Packed red blood cells and serum were used for evaluating different biochemical parameters.@*RESULTS@#Four weeks administration of aqueous extract of E. officinalis improved oral glucose tolerance in type 2 rats and after 8 weeks it caused significant (P<0.007) reduction in fasting serum glucose level compared to 0 day. Triglycerides decreased by 14% but there was no significant change in serum ALT, creatinine, cholesterol and insulin level in any group. Furthermore, reduced erythrocyte malondialdehyde level showed no significant change (P<0.07) but reduced glutathione content was found to be increased significantly (P<0.05).@*CONCLUSIONS@#The aqueous extract of E. officinalis has a promising antidiabetic and antioxidant properties and may be considered for further clinical studies in drug development.