Detection of Diatraea Saccharalis in images using convolutional neural networks

Agricultural pests cause 20-40 per cent loss of global crop production every year as reported by the Food and Agriculture Organization [1,2]. Excessive usage of pesticides to manage pests leads to severe problems. Smart agriculture presents the best option for farmers to apply artificial intelligence (AI) techniques integrated with modern information and communication technology to manage these harmful insect pests. Artificial intelligence (AI) is a broad term encompassing Machine Learning (ML), deep learning, computer vision etc. The core part of AI is Machine Learning (ML). Applications of AI in agricultural entomology are helpful in taxonomic studies, ecological studies and pest management. In this chapter, main focus is on AI usage in pest management through pest detection, monitoring, prediction and identification thereby helping in timely pest management. Several applications such as Plantix, Leaf-Byte, Bioleaf, Cotton Ace, Apizoom etc have been developed to diagnose and identify insect pests to manage them. Some of the important usage of AI in pest management discussed in the chapter are as follows: Chen et al. [3] developed an AIoT Based Smart Agricultural System for Tessaratoma papillosa (lychee giant stink bug) detection with 90 per cent accuracy. Karar et al. [4] developed a mobile application for the detection of five groups of insect pests viz., aphids, leaf hoppers, flax budworm, flea beetles, and red spider mites with 99.0 per cent accuracy for all tested pest images. As monitoring the insect pests is a crucial component in pheromone-based pest management systems, Ding and Taylor [5] developed an automatic moth detection method using AI with images collected from pheromone traps for timely pest management unlike the conventional counting methods. Liu et al. [6] using artificial intelligence developed an autonomous robotic vehicle in natural farm scene for the recognition of pyralidae insects with 94.3 per cent recognition accuracy for effective management of pyralidae insects in the farm. Potamitis and Rigakis [7] developed a smart trap for automatic remote monitoring of Rhynchophorus ferrugineus (Red palm weevil) to take necessary steps for controlling it based on ETLs. Selvaraj et al. [8] developed a model based on AI for banana diseases and pests detection with significant high success rate which is useful for early disease and pest detection. Hence, integrating artificial intelligence with Entomology will help in effective & timely management and forecasting of pests and diseases.

Deep learning algorithms, such as convolutional neural networks (CNNs), have been widely studied and applied in various fields including agriculture. Agriculture is the most important source of food and income in human life. In most countries, the backbone of the economy is based on agriculture. Pests are one of the major challenges in crop production worldwide. To reduce the overall production and economic loss from pests, advancement in computer vision and artificial intelligence may lead to early and small pest detection with greater accuracy and speed. In this paper, an approach for early pest detection using deep learning and convolutional neural networks has been presented. Object detection is applied on a dataset with images of thistle caterpillars, red beetles, and citrus psylla. The input dataset contains 9875 images of all the pests under different illumination conditions. State-of-the-art Yolo v3, Yolov3-Tiny, Yolov4, Yolov4-Tiny, Yolov6, and Yolov8 have been adopted in this study for detection. All of these models were selected based on their performance in object detection. The images were annotated in the Yolo format. Yolov8 achieved the highest mAP of 84.7% with an average loss of 0.7939, which is better than the results reported in other works when compared to small pest detection. The Yolov8 model was further integrated in an Android application for real time pest detection. This paper contributes the implementation of novel deep learning models, analytical methodology, and a workflow to detect pests in crops for effective pest management.

Deep learning (DL) has great influence on large parts of science and increasingly established itself as an adaptive method for new challenges in the field of Earth observation (EO). Nevertheless, the entry barriers for EO researchers are high due to the dense and rapidly developing field mainly driven by advances in computer vision (CV). To lower the barriers for researchers in EO, this review gives an overview of the evolution of DL with a focus on image segmentation and object detection in convolutional neural networks (CNN). The survey starts in 2012, when a CNN set new standards in image recognition, and lasts until late 2019. Thereby, we highlight the connections between the most important CNN architectures and cornerstones coming from CV in order to alleviate the evaluation of modern DL models. Furthermore, we briefly outline the evolution of the most popular DL frameworks and provide a summary of datasets in EO. By discussing well performing DL architectures on these datasets as well as reflecting on advances made in CV and their impact on future research in EO, we narrow the gap between the reviewed, theoretical concepts from CV and practical application in EO.

Agricultural production faces significant challenges due to pests and diseases, causing substantial losses in crop yield globally. Traditional methods of pest and disease management are often manual, time-consuming, and prone to errors. In recent years, the integration of artificial intelligence (AI) techniques, particularly deep learning algorithms, with modern information and communication technology has shown promising results in addressing these challenges. This paper presents a comprehensive review of recent advancements in applying deep learning for detecting and classifying agricultural pests, diseases, and weeds. Various deep learning models, including convolutional neural networks (CNNs) such as Faster R-CNN, InceptionV3, DenseNet, and AlexNet, have been explored for their efficacy in identifying pests, diseases, and weeds in crops. While models like Faster R-CNN, InceptionV3, and DenseNet have demonstrated high accuracy rates ranging from 78.71% to 99.62% in classification tasks across different datasets and crops, the AlexNet architecture has also shown promising results in certain applications within agricultural image analysis. Additionally, the development of lightweight CNN architectures and the fusion of deep features with traditional handcrafted features have further enhanced the accuracy and efficiency of detection systems. Furthermore, the review discusses challenges and future research directions in the field, emphasizing the importance of large-scale datasets, model optimization, and real-time applications for practical implementation in agriculture. Overall, the findings highlight the potential of deep learning technologies, including models like AlexNet, in revolutionizing pest and disease management practices, leading to improved crop yield, food security, and sustainable agriculture. Keywords: Agricultural pests, plant diseases, deep learning, convolutional neural networks (CNNs), AlexNet, classification

As technology continues to develop, computer vision (CV) applications are becoming increasingly widespread in the intelligent transportation systems (ITS) context. These applications are developed to improve the efficiency of transportation systems, increase their level of intelligence, and enhance traffic safety. Advances in CV play an important role in solving problems in the fields of traffic monitoring and control, incident detection and management, road usage pricing, and road condition monitoring, among many others, by providing more effective methods. This survey examines CV applications in the literature, the machine learning and deep learning methods used in ITS applications, the applicability of computer vision applications in ITS contexts, the advantages these technologies offer and the difficulties they present, and future research areas and trends, with the goal of increasing the effectiveness, efficiency, and safety level of ITS. The present review, which brings together research from various sources, aims to show how computer vision techniques can help transportation systems to become smarter by presenting a holistic picture of the literature on different CV applications in the ITS context.

We are pleased to present the first issue of Agricultural and Forest Entomology. This journal joins the other entomological journals of the Royal Entomological Society published by Blackwell Science, namely Ecological Entomology, Medical and Veterinary Entomology, Physiological Entomology, Insect Molecular Biology and Systematic Entomology. We look forward to working with a large international Editorial Board and Blackwell Science to produce Agricultural and Forest Entomology. Why launch a journal devoted to Agricultural and Forest Entomology? The problems caused by insect pests to agriculture and forestry have been with us as long as crops have been grown, but as both human populations and the consequential demand for food, timber and other products rise, the potentially damaging impact of pests has never been greater. Our ability to deal with insect pests has also increased as research on insect biology, systematics, physiology and ecology has provided a basis for managing them. The development, amongst other things, of insecticides, resistant crop varieties and biological control techniques has also led to dramatic improvements in the management of insect pests. However, insect pests are still a major obstacle to the harvesting of natural resources. This is partly due to the emergence of new pests, particularly as a result of the growth of non-native crops and the accidental introduction of non-native insects. There are many well-known cases of the latter problem, such as the gypsy moth (discussed here by Sharov et al. ) and the winter moth in North America. These and many other insect pests have been introduced from Europe. Less commonly, native insects become pests of non-native crop species but the result can be just as devastating, as illustrated by the damage caused by the pine beauty moth to lodgepole pine (a North American conifer) in the UK. Although new pest problems are constantly appearing, many ‘old’ pests continue to damage agricultural and forest crops. Papers on two of these pests appear in this issue: spruce budworm Choristoneura fumiferana (Cappuccino, Houle & Stein) and mahogany shoot borer Hypsipyla grandella (Newton et al. ). Despite considerable research aimed at managing these insects, the spruce budworm is still perhaps the most destructive forest pest in North America (Cappuccino et al. ), and mahogany shoot borers (Hypsipyla spp.) are arguably the most serious pests of tropical timber tree species. The continual threat posed by insect pests means that research on Agricultural and Forest Entomology has never been more important. Clearly there is a need for research on the biology of novel pests (e.g. Sage et al. ) and their impact (e.g. Stadler & Michalzik). To avoid the mistakes of the past and develop sound management of pests in the future, we need to know more about the ecology of most pest species. Research on both novel approaches to controlling pests and the possible side-effects of different methods of pest control (e.g. Thomas & Meats) is required. Agricultural and Forest Entomology will publish the best papers on these topics. Agricultural and Forest Entomology will publish papers relevant to the control of insect and other arthropod pests, including papers on biology, ecology, impact and management of pests of forest, agricultural and horticultural pests. Papers on the management of insect pests will include the development and use of techniques such as plant resistance, pheromones, biological control, the economics of pest control, and the use of silvicultural and crop management techniques for managing pests. Papers on the biology and ecology of insect pests (and their natural enemies) underpinning the development of pest management will also be published in Agricultural and Forest Entomology (e.g. Dickens; Smits & Larsson; Way et al. ). Papers relating to insect pests and their management in agricultural and forest crops in all parts of the world, including the tropics, will be welcome. We sincerely invite you to submit manuscripts to Agricultural and Forest Entomology. Scientific reports, short critical comments and longer reviews will all be considered (but please consult the Editors before submitting a longer review). If you have any comments, queries or suggestions please do not hesitate to contact the Editors or a member of the Editorial Board.

Chemically ripened mangoes and bananas are increasingly common worldwide and pose significant health risks due to the presence of carcinogens and other harmful substances. Owing to their cumbersome processes, the existing gold standard laboratory-based techniques for discriminating naturally or chemically ripened fruit often face challenges. This study aims to overcome these limitations by developing an onsite device specifically designed to detect chemically ripened mangoes and bananas to provide a faster and more cost-effective solution. This research uses advanced computer vision (CV) and deep learning (DL) techniques to detect and analyze chemically ripened mangoes and bananas. This research work employed several models, including K-nearest neighbor (KNN), random forest, support vector machine (SVM), convolutional neural networks (CNNs), and regional CNNs. In this study, the authors created their own real-time dataset for both naturally and chemically ripened mangoes and bananas. The proposed deep learning and machine learning models were trained and tested on a custom dataset of both fruits mango and banana images to discriminate chemically ripened fruits in an effective manner. Among all the models, the CNN achieved the highest accuracy of 93.24% and 96.25%, demonstrating its superior capability for this application. To scale up this approach, the authors implemented the system in real time via the Raspberry Pi board and a Pi camera. This prototype was instrumental for the authors to capture live images of fruits and process them via trained models to detect chemically ripened fruits. This approach enables efficient and accurate real-time detection, making this system feasible for practical applications. This work has the potential to leverage CV and DL techniques to combat fruit adulteration, providing a reliable and automated solution for ensuring food safety. The findings of this work infer that a CNN can accurately detect adulterated fruits, making it a promising tool for future developments in this field. Graphical Abstract

Artificial Intelligence for Computer Vision in Surgery: A Call for Developing Reporting Guidelines.

In June 1997, 1900 self-administered surveys were sent to Iowa alfalfa (Medicago sativa L.) producers. The sample was split into early adopter and general populations. Early adopters were those producers who purchased potato leafhopper (Empoasca fabae (Harris)] resistant alfalfa seed during the first year it was available commercially. The general population was randomly selected and no producer was present in both populations. Producers were asked questions on pest perceptions, management practices, and perceptions of leafhopper-resistant alfalfa. Seven hundred forty-seven usable surveys were returned. Results showed that the early adopter population farmed larger production systems and produced more alfalfa per acre than the general population. Overall, producers from the early adopter population had greater fundamental knowledge of serious alfalfa pests than producers from the general population. The potato leafhopper was reported as the most important pest in both populations. Producers seemed confused between the injury symptoms of potato leafhopper and alfalfa weevil [Hypera postica (Gyllenhal)] injury, and it was apparent that leafhopper injury symptoms were often confused with drought stress. Positive relationships were found between the frequency of scouting, the frequency of insecticide use, and alfalfa yield. The majority of producers expected leafhoppers to avoid resistant alfalfa varieties, resulting in smaller pest infestations. Adoption of leafhopper-resistant alfalfa probably will be limited by producers' knowledge of previous yield loss from this pest. Results of this survey will be used to develop improved educational materials on insect pest management in alfalfa.

Imperfections or defects inevitably occur in images due to inexperienced photographers, inadequate methods of preservation, or even some deliberate hacking. Image restoration or completion has been performed using various manual methods in the past, be it being drawn by artists based on their creativity or deleting noise and blur effects using software like Photoshop. On a large scale, manual image completion is infeasible and has quite a lot of limitations. Modern advancements in Computer Vision and Deep Learning have allowed man to automate such tasks with high efficiency. Manual restoration usually relies on prior experience in the subject and sometimes even creativity to reconstruct the image based on the artist's imagination. At the same time, deep learning produces excellent results given enough training data. Deep learning methods can improvise and generalize better too and hence outperform the traditional manual methods. In this project, image completion is performed using 2 Deep Learning models - Convolutional Neural Networks(CNN) and Generative Adversarial Networks(GAN). Adversarial Networks have been proven to be very handy in image to image translation tasks and image reconstruction and hence this is explored widely. Both Deep Convolutional GANs as well as Conditional GANs are used for this task and their respective performances are compared for the above task.

Detecting invertebrate pests on crops at early stages is essential for pest management. Traditionally, traps were used to sample pests and then human experts undertook classification and counting to estimate the levels of infestation, which is subjective, error-prone and labour intensive. Recently, semi-automatic pest detection is possible by using computer vision technologies to classify and count pest samples in laboratories or insect traps, however, the decision made by the laboratory-based or trap-based approaches are still too late for more optimised pest management decisions. Today, precision agriculture needs detection of pests on crops so that real-time actions can be taken or optimised decision can be made based on accurate information of time and location pest occurs. In this study, we used computer vision and machine learning technologies to detect invertebrates on crops in the field. We first evaluated the performances of the state-of-art convolutional neural networks (CNNs) and proposed a standard training pipeline. Facing the challenge of rapidly developing comprehensive training data, we used a novel method to generate a virtual database which was successfully used to train a deep residual CNN with an accuracy of 97.8% in detecting four species of pests in farming environments. The proposed method can be applied to a robotic system for proximal detection of invertebrate pests on crops in real-time.

Computer vision is the multidisciplinary domain extracts and analyses digital images in an automated manner. The application of computer vision is widespread and it ranges from agriculture to robotics. At present, computer vision adopts the concept of machine learning to build a model and solves classification problems. However, this technique becomes inefficient when it is directly applied to digital images as it ignores the structure and compositional nature of the images. Deep Convolutional Neural Network (CNN) acts as the best solution to traditional computer vision approaches as it learns to extract features from the raw images along with the classification process. In this paper, we present a deep learning based solution to computer vision problem. First, we define a CNN based approach to learn and extract features from the real time videos. Next, an extended linear support vector machine (SVM) classifier is used for object classification processes. Thus the proposed method make use of the combinational approach of the deep learning and machine learning to solve computer vision problems. Since deep CNN are massively parallel algorithms the application of CNN techniques with GPU forms the effective solution for computer vision problems. The experimental results are evaluated in terms performance, accuracy and simplicity measures.

Systems subjected to continuous operation are exposed to different failure mechanisms such as fatigue, corrosion, and temperature-related defects, which makes inspection and monitoring their health paramount to prevent a system suffering from severe damage. However, visual inspection strongly depends on a human being’s experience, and so its accuracy is influenced by the physical and cognitive state of the inspector. Particularly, civil infrastructures need to be periodically inspected. This is costly, time-consuming, labor-intensive, hazardous, and biased. Advances in Computer Vision (CV) techniques provide the means to develop automated, accurate, non-contact, and non-destructive inspection methods. Hence, this paper compares two different approaches to detecting cracks in images automatically. The first is based on a traditional CV technique, using texture analysis and machine learning methods (TA + ML-based), and the second is based on deep learning (DL), using Convolutional Neural Networks (CNN) models. We analyze both approaches, comparing several ML models and CNN architectures in a real crack database considering six distinct dataset sizes. The results showed that for small-sized datasets, for example, up to 100 images, the DL-based approach achieved a balanced accuracy (BA) of ∼74%, while the TA + ML-based approach obtained a BA > 95%. For larger datasets, the performances of both approaches present comparable results. For images classified as having crack(s), we also evaluate three metrics to measure the severity of a crack based on a segmented version of the original image, as an additional metric to trigger the appropriate maintenance response.

The rapid advancement of Artificial Intelligence (AI), Internet of Things (IoT) technologies, and predictive analytics is revolutionizing pest management strategies in agriculture. This chapter explores the integration of these cutting-edge technologies to develop intelligent, data-driven systems for pest detection, monitoring, and control. AI-powered computer vision systems, combined with IoT sensors and machine learning models, enable real-time monitoring of pest populations and environmental conditions, significantly improving the accuracy and efficiency of pest control interventions. Predictive analytics, leveraging historical and real-time data, further enhances these systems by forecasting pest outbreaks, allowing for proactive and targeted pest management. The chapter examines various real-world applications, including precision pest control in large-scale agriculture, greenhouses, vineyards, and field crops, where these integrated systems have successfully minimized chemical use, reduced environmental impact, and optimized resource allocation. Additionally, the challenges of scaling IoT sensor networks and the complexities of system integration are discussed, alongside potential solutions for widespread adoption. The future of pest management lies in the seamless fusion of AI, IoT, and predictive analytics, offering a sustainable, autonomous, and precision-driven approach to pest control. This chapter provides valuable insights for researchers, practitioners, and policymakers seeking to enhance pest management strategies in modern agriculture.

Artificial intelligent and application of computer vision are an exciting topic in last few years, and its key for many real time applications like video summarization, image retrieval and image classifications. One of the most trend method in deep learning is a convolutional neural network, used for many applications of image processing and computer vision. In this work convolutional neural networks CNN model proposed for color image classification, the proposed model build using MATLAB tools of deep learning. In addition, the suggested model tested on three different datasets, with different size. The proposed model achieved highest result of accuracy, precision and sensitivity with the largest dataset and it was as following: accuracy is 0.9924, precision is 0.9947 and sensitivity is 0.9931, compare with other models.