Alternaria leaf spot, Brown spot, Mosaic, Grey spot, and Rust are five common types of apple leaf diseases that severely affect apple yield. However, the existing research lacks an accurate and fast detector of apple diseases for ensuring the healthy development of the apple industry. This paper proposes a deep learning approach that is based on improved convolutional neural networks (CNNs) for the real-time detection of apple leaf diseases. In this paper, the apple leaf disease dataset (ALDD), which is composed of laboratory images and complex images under real field conditions, is first constructed via data augmentation and image annotation technologies. Based on this, a new apple leaf disease detection model that uses deep-CNNs is proposed by introducing the GoogLeNet Inception structure and Rainbow concatenation. Finally, under the hold-out testing dataset, using a dataset of 26,377 images of diseased apple leaves, the proposed INAR-SSD (SSD with Inception module and Rainbow concatenation) model is trained to detect these five common apple leaf diseases. The experimental results show that the INAR-SSD model realizes a detection performance of 78.80% mAP on ALDD, with a high-detection speed of 23.13 FPS. The results demonstrate that the novel INAR-SSD model provides a high-performance solution for the early diagnosis of apple leaf diseases that can perform real-time detection of these diseases with higher accuracy and faster detection speed than previous methods.

/pdf/real-time-detection-of-apple-leaf-diseases-using-deep-2x9rcpb8gd.pdf

Real-Time Detection of Apple Leaf Diseases Using Deep Learning Approach Based on Improved Convolutional Neural Networks

Plant diseases affect the growth of their respective species, therefore their early identification is very important. Many Machine Learning (ML) models have been employed for the detection and classification of plant diseases but, after the advancements in a subset of ML, that is, Deep Learning (DL), this area of research appears to have great potential in terms of increased accuracy. Many developed/modified DL architectures are implemented along with several visualization techniques to detect and classify the symptoms of plant diseases. Moreover, several performance metrics are used for the evaluation of these architectures/techniques. This review provides a comprehensive explanation of DL models used to visualize various plant diseases. In addition, some research gaps are identified from which to obtain greater transparency for detecting diseases in plants, even before their symptoms appear clearly.

https://www.mdpi.com/2223-7747/8/11/468/pdf

Plant Disease Detection and Classification by Deep Learning.

Plant leaf disease classification using EfficientNet deep learning model

Fungal diseases not only influence the economic importance of the plants and its products but also abate their ecological prominence. Mango tree, specifically the fruits and the leaves are highly affected by the fungal disease named as Anthracnose. The main aim of this paper is to develop an appropriate and effective method for diagnosis of the disease and its symptoms, therefore espousing a suitable system for an early and cost-effective solution of this problem. Over the last few years, due to their higher performance capability in terms of computation and accuracy, computer vision, and deep learning methodologies have gained popularity in assorted fungal diseases classification. Therefore, for this paper, a multilayer convolutional neural network (MCNN) is proposed for the classification of the Mango leaves infected by the Anthracnose fungal disease. This paper is validated on a real-time dataset captured at the Shri Mata Vaishno Devi University, Katra, J&K, India consists of 1070 images of the Mango tree leaves. The dataset contains both healthy and infected leaf images. The results envisage the higher classification accuracy of the proposed MCNN model when compared to the other state-of-the-art approaches.

/pdf/multilayer-convolution-neural-network-for-the-classification-1b4ioub2xp.pdf

Multilayer Convolution Neural Network for the Classification of Mango Leaves Infected by Anthracnose Disease

This paper presents a hybrid wearable sensor network system towards the Internet of Things (IoT) connected safety and health monitoring applications. The system is aimed at improving safety in the outdoor workplace. The proposed system consists of a wearable body area network (WBAN) to collect user data and a low-power wide-area network (LPWAN) to connect the WBAN with the Internet. The wearable sensors in the WBAN are exerted to measure the environmental conditions around the subject using a Safe Node and monitor the vital signs of the subject using a Health Node. A standalone local server (gateway), which can process the raw sensor signals, display the environmental and physiological data, and trigger an alert if any emergency circumstance is detected, is designed within the proposed network. To connect the gateway with the Internet, an IoT cloud server is implemented to provide more functionalities, such as web monitoring and mobile applications.

/pdf/an-internet-of-things-iot-network-system-for-connected-4rmsvgyohz.pdf

An Internet-of-Things (IoT) Network System for Connected Safety and Health Monitoring Applications

In the field of agricultural information, the automatic identification and diagnosis of maize leaf diseases is highly desired. To improve the identification accuracy of maize leaf diseases and reduce the number of network parameters, the improved GoogLeNet and Cifar10 models based on deep learning are proposed for leaf disease recognition in this paper. Two improved models that are used to train and test nine kinds of maize leaf images are obtained by adjusting the parameters, changing the pooling combinations, adding dropout operations and rectified linear unit functions, and reducing the number of classifiers. In addition, the number of parameters of the improved models is significantly smaller than that of the VGG and AlexNet structures. During the recognition of eight kinds of maize leaf diseases, the GoogLeNet model achieves a top - 1 average identification accuracy of 98.9%, and the Cifar10 model achieves an average accuracy of 98.8%. The improved methods are possibly improved the accuracy of maize leaf disease, and reduced the convergence iterations, which can effectively improve the model training and recognition efficiency.

Identification of Maize Leaf Diseases Using Improved Deep Convolutional Neural Networks

To meet the requirements of long range, a small amount of data transmission, low power, and low cost of the Internet of Things (IoT) in actual applications, a low-power wide-area network information monitoring approach based on NB-IoT and LoRa is proposed in this paper. This approach adopts a communication mode that contains a main node and multiple subnodes to adapt to the needs of large-scale information monitoring. Among them, the design of the main node utilizes NB-IoT communication technology, LoRa communication technology, and least recently used algorithm. The design of subnode utilizes LoRa communication technology, sensor technology, and optical-electric conversion technology. Finally, we design and implement a cloud service and computing system, using the Tencent Cloud server. The advantage of this approach is that the combination of NB-IoT and LoRa not only improves transmission distance but also reduces the operating costs of the WAN information monitoring system. In addition, using optical-electric conversion technology, the system can be self-powered. Combined with the principle of electric circuits, the power consumption of the system can also be reduced. Finally, by conducting a system test, LoRa communication distance experiment, NB-IoT communication, and other experiments, the communication distance in a complex environment is up to 1.6 km, the minimum working current is 2 mA, and the system communication packet loss rate is approximately 3%. The system runs stably and the collected data are accurate. The results show that the proposed approach can meet the requirements of data collection, transmission, storage, and calculation in a wide area.

A Low-Power Wide-Area Network Information Monitoring System by Combining NB-IoT and LoRa

The invention relates to obstacle avoiding methods for agricultural plant protection unmanned aerial vehicle (UAV)s under multiple obstacles and small obstacles in the spraying process and a UAV, and belongs to the field of UAV flight route planning. The invention aims to solve the problem that the UAV spraying flight route planning method in the prior art fails to select an obstacle avoiding route suitable for the maneuverability of UAVs and does not take into account the need for UAVs to fly back along the original route for spraying, which results in large overlapping and deficient spraying area. The obstacle avoiding method for agricultural plant protection UAVs under multiple obstacles in the spraying process comprises the following steps: connecting the start point and end point of a flight route to form a line segment; judging whether the line segment intersects with an obstacle circle; if the line segment intersects with the obstacle circle, generating a minimum turning circle which is tangent to the line segment and the obstacle at the same time, and determining a plurality of optional flight routes according to the line segment, the obstacle circle and the minimum turning circle; and selecting an optimal flight route according to fitness values calculated based on the lengths of flight routes and corresponding overlapping and deficient spraying areas. The methods are suitable for route planning in the spraying process of agricultural plant protection UAVs.

Obstacle avoiding methods for agricultural plant protection unmanned aerial vehicles under multiple obstacles and small obstacles in spraying process and unmanned aerial vehicle

The invention provides an agricultural plant protection unmanned aerial vehicle and an obstacle avoidance and spraying path planning method for the same, and relates to the field of route planning methods and control methods for unmanned aerial vehicles. The agricultural plant protection unmanned aerial vehicle and the novel obstacle avoidance and spraying path planning method aim to overcome the shortcoming that reduction of the lengths of existing paths and reduction of repeated spraying and missed spraying cannot be taken into account by obstacle avoidance algorithms in the prior art. The obstacle avoidance and spraying path planning method includes judging whether straight line sections intersect with obstacle circles or not; generating Dubins paths if the straight line sections intersect with the obstacle circles, determining all optional flight paths and calculating flight path lengths of first paths, the lengths of swerving path, and repeated spraying areas; repeatedly carrying out previous steps until all flight path lengths and repeated spraying areas are calculated. The straight line sections comprise starting points and end points of first spraying paths. The agricultural plant protection unmanned aerial vehicle and the obstacle avoidance and spraying path planning method have the advantages that genetic algorithms are additionally implemented in order to quickly find out the optimal paths, and the optimal paths can be quickly selected from the optional flight path; the obstacle avoidance and spraying path planning method is applicable to plant protection unmanned aerial vehicles.

Agricultural plant protection unmanned aerial vehicle and obstacle avoidance and spraying path planning method for same

The invention relates to an obstacle avoidance method of a single obstacle and a small obstacle in spraying process of an agricultural plant protection unmanned machine and the unmanned machine, belongs to the flying path planning domain of the unmanned machine. In order to solve the shortcomings that the spraying and flying path planning method of the unmanned machine in the prior art has not selected a suitable obstacle avoidance path for the maneuvering characteristics of the unmanned machine, and has not considered about the spraying of the unmanned machine at the original path so that the re-spraying and leaked-spraying areas are too large. The obstacle avoidance method of a single obstacle in spraying process of the agricultural plant protection unmanned machine includes steps of judging if the straight section of an original flying path is tangential to an obstacle circle; if it is, generating a plurality of minimum turning circles which are tangential to the straight section and the target obstacle circle at the same time; according to the straight section, the target obstacle circle and the minimum turning circle, confirming a plurality of optional flying paths; selecting the path with the highest evaluation value after the normalization processing from the selectable paths as the final spraying and flying path. The invention is applicable to the line planning during the spraying work of the agricultural plant protection unmanned machine.

Obstacle avoidance method of single obstacle and small obstacle in spraying process of agricultural plant protection unmanned machine and unmanned machine

Qiao Yue

Papers

Identification of Maize Leaf Diseases Using Improved Deep Convolutional Neural Networks

A Low-Power Wide-Area Network Information Monitoring System by Combining NB-IoT and LoRa

Obstacle avoiding methods for agricultural plant protection unmanned aerial vehicles under multiple obstacles and small obstacles in spraying process and unmanned aerial vehicle

Agricultural plant protection unmanned aerial vehicle and obstacle avoidance and spraying path planning method for same

Obstacle avoidance method of single obstacle and small obstacle in spraying process of agricultural plant protection unmanned machine and unmanned machine