Computers and Electronics in Agriculture最新文献

{"title":"Improved composite deep learning and multi-scale signal features fusion enable intelligent and precise behaviors recognition of fattening Hu sheep","authors":"Mengjie Zhang ,&nbsp;Yanfei Zhu ,&nbsp;Jiabao Wu ,&nbsp;Qinan Zhao ,&nbsp;Xiaoshuan Zhang ,&nbsp;Hailing Luo","doi":"10.1016/j.compag.2024.109635","DOIUrl":"10.1016/j.compag.2024.109635","url":null,"abstract":"<div><div>The integration of artificial intelligence and advanced sensing technologies can improve the intelligence and precision level of livestock management. This study focuses on fattening Hu sheep as the object of study, and aims to assess the effectiveness of integrating multi-scale biological signals with improved composite deep learning model in identifying and classifying behaviors of fattening Hu sheep. The multi-scale biological signals were collected using the respiratory sensor and the multi-dimensional posture sensor (composed of an accelerometers, gyroscope, and magnetometer), and then, after data processing, extracted signal features and used the dimensionality reduction method of principal component analysis (PCA). Attention-based particle swarm optimized convolution and long short-term memory (APSO-CALM) model was developed using the feature fused dataset, and its performance was compared with other models. The results showed that: (1) The multi-scale biological signals were analyzed and categorized into five distinct behaviors based on experimental records: feeding, rumination, mating, free movement and running. Each of these behaviors exhibits unique characteristics in their signal images. (2) PCA was utilized to reduce the dimensionality of the feature fused dataset of the multi-scale biological signals, preserving principal components with a cumulative contribution rate of 98 %. Among all components of the first and second contribution rates, except for a few individuals, there are significant differences (P &lt; 0.05) between the data of different behaviors of the same component. (3) The improved composite deep learning model, APSO-CALM, demonstrates significant advantages over single models in behavior recognition. Its accuracy, precision, recall, and F1 score are 95.0 %, 94.8 %, 94.5 %, and 94.6 %, respectively. By utilizing the APSO-CALM model, the drawbacks of individual models are mitigated, enhancing overall performance and overcoming the limitations of single model applications. This study effectively identified five behaviors of fattening Hu sheep, providing theoretical and practical basis for intelligent and precise management of fattening Hu sheep.</div></div>","PeriodicalId":50627,"journal":{"name":"Computers and Electronics in Agriculture","volume":"227 ","pages":"Article 109635"},"PeriodicalIF":7.7,"publicationDate":"2024-11-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142661723","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Benchmarking of monocular camera UAV-based localization and mapping methods in vineyards","authors":"Kaiwen Wang ,&nbsp;Lammert Kooistra ,&nbsp;Yaowu Wang ,&nbsp;Sergio Vélez ,&nbsp;Wensheng Wang ,&nbsp;João Valente","doi":"10.1016/j.compag.2024.109661","DOIUrl":"10.1016/j.compag.2024.109661","url":null,"abstract":"<div><div>UAVs equipped with various sensors offer a promising approach for enhancing orchard management efficiency. Up-close sensing enables precise crop localization and mapping, providing valuable a priori information for informed decision-making. Current research on localization and mapping methods can be broadly classified into SfM, traditional feature-based SLAM, and deep learning-integrated SLAM. While previous studies have evaluated these methods on public datasets, real-world agricultural environments, particularly vineyards, present unique challenges due to their complexity, dynamism, and unstructured nature.</div><div>To bridge this gap, we conducted a comprehensive study in vineyards, collecting data under diverse conditions (flight modes, illumination conditions, and shooting angles) using a UAV equipped with high-resolution camera. To assess the performance of different methods, we proposed five evaluation metrics: efficiency, point cloud completeness, localization accuracy, parameter sensitivity, and plant-level spatial accuracy. We compared two SLAM approaches against SfM as a benchmark.</div><div>Our findings reveal that deep learning-based SLAM outperforms SfM and feature-based SLAM in terms of position accuracy and point cloud resolution. Deep learning-based SLAM reduced average position error by 87% and increased point cloud resolution by 571%. However, feature-based SLAM demonstrated superior efficiency, making it a more suitable choice for real-time applications. These results offer valuable insights for selecting appropriate methods, considering illumination conditions, and optimizing parameters to balance accuracy and computational efficiency in orchard management activities.</div></div>","PeriodicalId":50627,"journal":{"name":"Computers and Electronics in Agriculture","volume":"227 ","pages":"Article 109661"},"PeriodicalIF":7.7,"publicationDate":"2024-11-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142661676","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Multi-Hop LoRa-based underground network for monitoring soil moisture in agriculture","authors":"Reinaldo Cotrim ,&nbsp;Flávio Assis ,&nbsp;Alexsandro dos Santos Brito ,&nbsp;Yslai Silva Peixouto ,&nbsp;Leandro Santos Peixouto","doi":"10.1016/j.compag.2024.109592","DOIUrl":"10.1016/j.compag.2024.109592","url":null,"abstract":"<div><div>Wireless underground sensor networks (WUSN) have gained attention due to the benefits they can bring to many application areas, in particular, to agriculture. However, designing and evaluating WUSNs is more complex than conventional over-the-air wireless networks, especially when the WUSNs have buried nodes. The study of the possibilities and limits of these networks is an active area of research. In this paper we describe a LoRa-based multi-hop WUSN for monitoring soil moisture for an application in agriculture being developed to investigate the behaviour of different species of <em>mamona</em> (<em>Ricinus communis L.</em>) under different soil moisture levels. We first evaluate the use of LoRa for underground-to-underground (UG2UG) communication links and show how different values of the main LoRa parameters affect the quality of these links. Based on the results, we designed a network whose topology is a set of lines of buried sensor nodes covering the whole application area. In this paper we describe the behaviour of one of these lines in a real setting in terms of packet delivery ratio and delay and we estimate the energy consumed for communication. Our protocol provides an inherent level of fault-tolerance by exploring the linear topology. In our experiments, a 100% message delivery ratio was achieved. Additionally, the maximum round-trip delay was less than 200 s. The network satisfies the application message transmission requirement of one message per hour per node by scheduling communication over the six sensor lines needed to cover the whole experiment area in a round-robin fashion. Our main contributions lie in the evaluation of different parameters of LoRa in underground communication and in the development and analysis of a multi-hop routing protocol for a network of buried nodes in a real setting. We are not aware of any other work that addresses these specific issues.</div></div>","PeriodicalId":50627,"journal":{"name":"Computers and Electronics in Agriculture","volume":"227 ","pages":"Article 109592"},"PeriodicalIF":7.7,"publicationDate":"2024-11-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142661725","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Deep learning-based classification of visual symptoms of bacterial wilt disease caused by Ralstonia solanacearum in tomato plants","authors":"J.P. Vásconez ,&nbsp;I.N. Vásconez ,&nbsp;V. Moya ,&nbsp;M.J. Calderón-Díaz ,&nbsp;M. Valenzuela ,&nbsp;X. Besoain ,&nbsp;M. Seeger ,&nbsp;F. Auat Cheein","doi":"10.1016/j.compag.2024.109617","DOIUrl":"10.1016/j.compag.2024.109617","url":null,"abstract":"<div><div>Classification of plant diseases based on computer vision is a multidisciplinary challenge that involves technical and data-related complexities. Artificial Intelligence (AI) has increasingly found its application in plant pathology, disease, and anomaly visual characterization. Specifically, Machine Learning (ML) and Deep Learning (DL) algorithms have proven to be highly effective for tasks such as plant disease classification, detection, diagnosis, and management. In this work, we present a comparative analysis of multiple DL models based on Convolutional Neural Networks (CNNs) for visual symptoms classification of the phytopathogen <em>Ralstonia solanacearum</em> in tomato plants. We demonstrate that by implementing DL classification algorithms based on CNNs, it is possible to identify <em>Ralstonia solanacearum</em> potentially affected plants. This was possible due to the main virulence factor of <em>Ralstonia solanacearum</em>, the exopolysaccharide (EPS), which obstructs the plant’s xylem limiting water absorption and consequently inducing visual wilting symptoms. For this, we implemented, trained, and evaluated fourteen different CNN-based models. We evaluated the models by using different metrics such as precision, recall, accuracy, specificity, and F1-score. The models that obtained the best accuracy results were MobileNet-v2 and Xception, with an accuracy of 97.7% for both models. The presented findings significantly contribute to the visual symptoms classification of <em>Ralstonia solanacearum</em> in tomato plants, which may contribute to the control of this disease and its spread to healthy crops or other susceptible hosts in the future.</div></div>","PeriodicalId":50627,"journal":{"name":"Computers and Electronics in Agriculture","volume":"227 ","pages":"Article 109617"},"PeriodicalIF":7.7,"publicationDate":"2024-11-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142661771","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Integrating masked generative distillation and network compression to identify the severity of wheat fusarium head blight","authors":"Zheng Gong,&nbsp;Chunfeng Gao,&nbsp;Zhihui Feng,&nbsp;Ping Dong,&nbsp;Hongbo Qiao,&nbsp;Hui Zhang,&nbsp;Lei Shi,&nbsp;Wei Guo","doi":"10.1016/j.compag.2024.109647","DOIUrl":"10.1016/j.compag.2024.109647","url":null,"abstract":"<div><div>Fusarium head blight (FHB) is a severe disease, with implications for both crop quality and safety. The inability to accurately and rapidly determine diseases severity has resulted in increasing grain loss and the pesticide expenses. Furthermore, the complexity of many current models presents challenges in their deployment and utilization. Thus, this study introduces an improved lightweight model for efficient and rapid assessment of FHB severity. Firstly, we collected 2650 wheat images with different severities in natural environments. Second, we refined and compressed RepGhostNet, replacing the original ReLU function with LeakyReLU and using the AdamW optimizer during training to enhance model accuracy. Third, using the strategy of masked generative distillation, we further improved the accuracy of SlimRepGhostNet while ensuring model lightweight. The MGD-SlimRepGhostNet achieved an accuracy of 94.58% and a frames per second (FPS) of 152.17. This represents a 4.34% increase in accuracy and a 21.17 increase in speed compared to the original RepGhostNet. Lastly, we have designed a WeChat mini program that achieves the performance of MGD-SlimRepGhostNet in real environments, highlighting its practicality. The proposed method effectively addresses the inaccuracies and labor-intensive associated with nature of traditional visual assessment methods deployed for evaluating FHB severity in wheat, while its rapid inference capability renders it highly suitable for deployment and application on mobile devices.</div></div>","PeriodicalId":50627,"journal":{"name":"Computers and Electronics in Agriculture","volume":"227 ","pages":"Article 109647"},"PeriodicalIF":7.7,"publicationDate":"2024-11-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142661681","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Development and evaluation of a dual-arm robotic apple harvesting system","authors":"Kyle Lammers ,&nbsp;Kaixiang Zhang ,&nbsp;Keyi Zhu ,&nbsp;Pengyu Chu ,&nbsp;Zhaojian Li ,&nbsp;Renfu Lu","doi":"10.1016/j.compag.2024.109586","DOIUrl":"10.1016/j.compag.2024.109586","url":null,"abstract":"<div><div>Harvesting labor is the single largest cost in apple production in the U.S. Increased cost and growing shortage of labor has forced the apple industry to seek automated harvesting solutions. Despite considerable progress in recent years, the existing robotic harvesting systems still fall short of performance expectations, lacking robustness and proving inefficient or overly complex for practical commercial deployment. In this paper, we present the development and evaluation of a new dual-arm robotic apple harvesting system. The system hardware mainly consists of a perception component, two four-degree-of-freedom manipulators, a centralized vacuum system, and a fruit handling and bin filling component designed for the collection and transportation of picked fruits. Synergistic functionalities for automated apple harvesting were achieved through the development of software algorithms. In particular, an updated perception system based on dual-laser scanning was proposed to enable sequential localization of apples for the dual-arm robotic system. A sophisticated planning scheme was devised to coordinate the movement of the two manipulators, allowing them to approach the fruit effectively and share a centralized vacuum system for efficient fruit detachment. The robotic system has been evaluated through field trials in a challenging apple orchard with complex, dense canopy, and it achieved 60% successful picking rate. The dual-arm coordination algorithm resulted in 9% to 34% harvest time improvements, compared to the 1-arm robotic system design. The new dual-arm robotic system is compact in design and dexterous in movement, and with further improvements in hardware and software, it holds great potential for providing a commercially viable harvesting automation solution for the apple industry</div></div>","PeriodicalId":50627,"journal":{"name":"Computers and Electronics in Agriculture","volume":"227 ","pages":"Article 109586"},"PeriodicalIF":7.7,"publicationDate":"2024-11-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142661728","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Effect of hydraulic configuration on lettuce growth in hydroponic bed using Deep water culture technique (DWC)","authors":"Carlos J. Cortés ,&nbsp;Nelson O. Moraga ,&nbsp;Constanza Jana ,&nbsp;Germán E. Merino","doi":"10.1016/j.compag.2024.109634","DOIUrl":"10.1016/j.compag.2024.109634","url":null,"abstract":"<div><div>Experiments and computational modeling were developed to determine the effect of different types of hydraulic configurations on water quality variables to improve growth of lettuce in hydroponic beds. The variants in the hydraulic configurations consider water recirculation in hydroponic modules using Deep Water Culture technique (DWC), for continuous (CWF) and pulsatile water flow (PWF) using either one or three water flow inlets (TWF). These data were used to generate fluid mechanics and heat transfer models for the described hydraulic configurations to assess the effect of the hydraulic configuration on lettuce growth. The results obtained from the mathematical model by the finite volume method allowed to explain the influence of water flow and temperature on the rate of growing for lettuce during summer and autumn in the southern hemisphere. The main findings obtained from the hybrid numerical – experimental model to achieve high lettuce yield were that the number of water inlets has an effect on influenced nutrient transport and water quality variation, where the variant with three water inlets (TWF), and the climatic condition for autumn achieve better plant growth performance than summer. Computational modelling of fluid mechanics and heat transfer allowed to predict the variation of water quality variables in DWC bed, being a suitable technique with a high potential for achieving new accurate agriculture standards.</div></div>","PeriodicalId":50627,"journal":{"name":"Computers and Electronics in Agriculture","volume":"227 ","pages":"Article 109634"},"PeriodicalIF":7.7,"publicationDate":"2024-11-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142661765","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Enhanced detection of mango leaf diseases in field environments using MSMP-CNN and transfer learning","authors":"Yi-Chen Chen ,&nbsp;Jen-Cheng Wang ,&nbsp;Mu-Hwa Lee ,&nbsp;An-Chi Liu ,&nbsp;Joe-Air Jiang","doi":"10.1016/j.compag.2024.109636","DOIUrl":"10.1016/j.compag.2024.109636","url":null,"abstract":"<div><div>Mango trees affected by various diseases often exhibit distinctive leaf symptoms. Accurate and timely diagnosis is crucial for mango cultivation. Deep learning algorithms provide a viable solution for precisely detection of mango leaf diseases. However, two main challenges exist: environmental interference and the difficulty of collecting leaf image data from the field. To address these challenges, this study introduces a multi-scale and multi-pooling convolutional neural network (MSMP-CNN) model. The proposed model undergoes a pre-training phase, followed by transfer learning and fine-tuning, and ultimately focuses on identifying mango leaf diseases using real-world images. This model exhibits outstanding performance in identifying various mango leaf diseases. The model achieved an accuracy of 95 % on its own. After being enhanced by transfer learning and find-tuning, the model achieved an impressive accuracy of 98.5 %. To compare the classification performance with and without transfer learning and fine-tuning, t-distributed stochastic neighbor embedding (t-SNE) plots were used. Class activation mapping (CAM) heatmaps were also utilized to highlight class-specific regions of images, helping verify whether the model focused on the appropriate parts of the image for disease identification. These findings underscore the strong potential of the model combining with transfer learning and fine-tuning to advance mango leaf disease detection. In the future, the proposed model will evolve into a real-time, precise diagnostic system for mango leaf diseases, thereby transforming mango cultivation management from precision farming to smart agriculture.</div></div>","PeriodicalId":50627,"journal":{"name":"Computers and Electronics in Agriculture","volume":"227 ","pages":"Article 109636"},"PeriodicalIF":7.7,"publicationDate":"2024-11-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142661722","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Evaluating agriculture 4.0 decision support systems based on hyperbolic fuzzy-weighted zero-inconsistency combined with combinative distance-based assessment","authors":"Abdullah Alamoodi ,&nbsp;Salem Garfan ,&nbsp;Muhammet Deveci ,&nbsp;O.S. Albahri ,&nbsp;A.S. Albahri ,&nbsp;Salman Yussof ,&nbsp;Raad Z. Homod ,&nbsp;Iman Mohamad Sharaf ,&nbsp;Sarbast Moslem","doi":"10.1016/j.compag.2024.109618","DOIUrl":"10.1016/j.compag.2024.109618","url":null,"abstract":"<div><div>Agriculture 4.0 plays a crucial role in shaping sustainable cities and societies by revolutionizing urban food systems. By incorporating advanced technologies like precision farming, vertical gardening, and data analytics, Agriculture 4.0 improves local food production, reduces food transportation, and optimizes resource utilization. This paper introduces an innovative approach using Multi-Criteria Decision Making (MCDM) to assess Agriculture 4.0 Decision Support Systems (ADSS), contributing significantly to the selection of optimal systems that can drive sustainability in smart agriculture. The novelty of this research lies in developing a comprehensive evaluation framework that extends the hyperbolic fuzzy-weighted zero-inconsistency method for criteria weighting, combined with the combinative distance-based assessment method for benchmarking ADSS. The assessment matrix evaluates 13 ADSS across eight key criteria, including “<em>accessibility</em>,” “<em>re-planning</em>,” “<em>expert knowledge</em>,” “<em>interoperability</em>,” “<em>scalability</em>,” “<em>uncertainty and dynamic factors</em>,” “<em>prediction and forecast</em>,” and “<em>historical data analysis</em>”. Results from the hyperbolic fuzzy-weighted zero-inconsistency approach highlight “<em>re-planning</em>” (<em>0.143</em>) and “<em>prediction and forecast</em>” (<em>0.140</em>) as the most significant criteria, while “<em>expert knowledge</em>” ranked lowest (<em>0.113</em>). In the combinative distance-based assessment, the system labelled “OCCASION” achieved the highest score (<em>3.843</em>), positioning it as the most favourable ADSS, whereas the “MOLP-based beef supply chain” system scored lowest <em>(−3.519</em>). Sensitivity analysis, conducted using varying sets of weights, confirms the robustness and reliability of the proposed approach. This research provides a powerful decision-making tool that can guide stakeholders in selecting the best ADSS, ultimately promoting sustainability and resource optimization in Agriculture 4.0. The findings have important implications for farmers, agribusiness, and smart agriculture, demonstrating the potential of the methodology to enhance decision-making processes in a critical sector.</div></div>","PeriodicalId":50627,"journal":{"name":"Computers and Electronics in Agriculture","volume":"227 ","pages":"Article 109618"},"PeriodicalIF":7.7,"publicationDate":"2024-11-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142661726","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A monochrome pipelined HMI system for foodborne microorganisms testing","authors":"Jia-Yong Song ,&nbsp;Ze-Sheng Qin ,&nbsp;Chang-Wen Xue ,&nbsp;Li-Feng Bian ,&nbsp;Chen Yang","doi":"10.1016/j.compag.2024.109650","DOIUrl":"10.1016/j.compag.2024.109650","url":null,"abstract":"<div><div>Hyperspectral microscopy imaging (HMI) is an efficient and non-destructive method to detect microbial contaminants in food, as it can provide both spatial morphology and spectral signature. Aims at reducing thermal effect, low cost, and improving spectral resolution in testing, a pipeline-operated LEDs monochromatic illumination mode is proposed, which integrates the design concepts of both grating-based and LED-based HMI systems. By design of the LED set, shared grating monochromatic optical path, and coordinated control system, an HMI system has been developed that could obtain the hyperspectral data cube with 101 bands in 400–700 <em>nm</em>. Hyperspectral datasets of three species of Aspergillus are prepared using the prototype, and efficient results have been achieved in the training and testing of classical classification algorithms (1D-CNN (97.33 %), k-NN (96.33 %), SVM (97.67 %) and ResNet-18 (95.67 %)). The results demonstrate that the proposed monochromatic illumination mode and associated system are potential detection solutions for foodborne microbial contaminants with low-cost and high-accurate.</div></div>","PeriodicalId":50627,"journal":{"name":"Computers and Electronics in Agriculture","volume":"227 ","pages":"Article 109650"},"PeriodicalIF":7.7,"publicationDate":"2024-11-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142661770","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"农林科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}