[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"item-677":3,"related-677":64},{"id":4,"title":5,"url":6,"summary":7,"summary_zh":8,"content":9,"source_name":10,"source_url":8,"published_at":11,"category":12,"cover_url":8,"hotness":13,"is_selected":14,"score":15,"score_detail":16,"sources":23,"tags":25,"primary_topic":30,"secondary_topics":36,"classification":37,"search_phrases":58,"slug":61,"view_count":62,"doi":8,"paper":8,"created_at":63},677,"Computer Vision from Tea Cultivation to Quality Evaluation(MDPI Foods 2026-08-17)","https:\u002F\u002Fwww.mdpi.com\u002F2304-8158\u002F15\u002F16\u002F2864","茶叶栽培到品质评估的计算机视觉综述。全面比较视觉技术(RGB、高光谱、近红外、热成像、LiDAR、UAV)并建立任务导向的算法选择框架。小样本或近线性问题传统机器学习(SVM、PLSR)依然有效;非结构化田间任务深度学习表现优越:害虫检测准确率超97%,茶芽RGB图像检测96.8%,高光谱预测氮含量R²>0.90、茶多酚R²高达0.925。算法选择需进一步按任务粒度区分:轻量级CNN平衡边缘部署速度与精度(16fps);YOLO系列检测器在移动平台实现实时定位(93.1%准确率、24ms\u002F目标)。",null,"Logical Operator Operator\n\nSearch Text\n\nSearch Type\n\n_add\\_circle\\_outline_\n\n_remove\\_circle\\_outline_\n\n[![Image 1: foods-logo](https:\u002F\u002Fpub.mdpi-res.com\u002Fimg\u002Fjournals\u002Ffoods-logo.png?a8f99d50b4c10c98)](https:\u002F\u002Fwww.mdpi.com\u002Fjournal\u002Ffoods)\n\n## Article Menu\n\nFont Type:\n\n_Arial_ _Georgia_ _Verdana_\n\nFont Size:\n\nAa Aa Aa\n\nLine Spacing:\n\n__ __ __\n\nColumn Width:\n\n__ __ __\n\nBackground:\n\nOpen Access Review\n\nby \nZunren Chen\n\n 1,†[](mailto:c2021010@fjtcm.edu.cn), \nJinfeng Wang\n\n 2,†[](mailto:18674272913@163.com), \nYilan Sun\n\n 2[](mailto:yilansun@126.com)[![Image 2: ORCID](https:\u002F\u002Fpub.mdpi-res.com\u002Fimg\u002Fdesign\u002Forcid.png?0465bc3812adeb52?1786617101)](https:\u002F\u002Forcid.org\u002F0009-0008-2996-1733), \nJie Pang\n\n 2[](mailto:pang3721941@163.com)[![Image 3: ORCID](https:\u002F\u002Fpub.mdpi-res.com\u002Fimg\u002Fdesign\u002Forcid.png?0465bc3812adeb52?1786617101)](https:\u002F\u002Forcid.org\u002F0000-0002-7061-5233), \nWei Xin\n\n 3,4[](mailto:xinwei@wuyiu.edu.cn), \nQinhua Zhang\n\n 5,*[](mailto:m15716056182@163.com) and \nJunling Zhou\n\n 6,*[](mailto:junling-zhou@fafu.edu.cn)\n\n1\n\nLibrary of Fujian University of Traditional Chinese Medicine, Fuzhou 350122, China\n\n2\n\nSchool of Food Science, Fujian Agriculture and Forestry University, Fuzhou 350002, China\n\n3\n\nCollege of Ecology and Resources Engineering, Wuyi University, Wuyishan 354300, China\n\n4\n\nFujian Key Laboratory of Big Data Application and Intellectualization for Tea Industry, Wuyi University, Nanping 529000, China\n\n5\n\nJinshan College, Fujian Agriculture and Forestry University, Fuzhou 350002, China\n\n6\n\nHaixia Institute of Science and Technology, Fujian Agriculture and Forestry University, Fuzhou 350002, China\n\n*\n\nAuthors to whom correspondence should be addressed.\n\n†\n\nThese authors contributed equally to this work.\n\nSubmission received: 23 June 2026 \u002F Revised: 7 August 2026 \u002F Accepted: 11 August 2026 \u002F Published: 17 August 2026\n\n## Abstract\n\nExisting reviews on AI in tea production are either agriculture-generic or limited to isolated tasks. This review thoroughly compares vision technologies (RGB, hyperspectral, near-infrared, thermal, Light Detection and Ranging (LiDAR), Unmanned Aerial Vehicle (UAV)) and establishes a task-oriented algorithm selection framework for the tea industry. For small-sample or near-linear problems, traditional machine learning (ML) (support vector machine (SVM); partial least squares regression (PLSR)) remains effective. For unstructured field tasks, deep learning achieves superior performance: pest detection accuracy exceeds 97%, tea bud detection reaches 96.8% with RGB images, and hyperspectral imaging predicts nitrogen content with R 2> 0.90 and tea polyphenols with R 2 up to 0.925. Algorithm choice further differentiates by task granularity: lightweight convolutional neural networks (CNNs) balance speed and accuracy for edge deployment at 16 fps; You Only Look Once (YOLO) series detectors enable real-time localization on mobile platforms at 93.1% accuracy, 24 ms per target. No single algorithm dominates all tea tasks; selection is a trade-off among accuracy, speed, data availability, and computational constraints. These findings outline a structured analysis of the challenges and pathways for transitioning computer vision (CV) from laboratory research toward field-deployable tools.\n\n## 1. Introduction\n\nTea, an ancient beverage originating in the East, is now the second most consumed drink globally after water [[1](https:\u002F\u002Fwww.mdpi.com\u002F2304-8158\u002F15\u002F16\u002F2864#B1-foods-15-02864)]. It constitutes a major global industry at the intersection of culture, health, and economics. According to the Food and Agriculture Organization (FAO), global tea production exceeded 6.5 million tons in 2022, with a trade value of approximately $17 billion USD (FAO, 2022), serving as an economic pillar in many rural and developing regions [[2](https:\u002F\u002Fwww.mdpi.com\u002F2304-8158\u002F15\u002F16\u002F2864#B2-foods-15-02864)]. On one hand, consumer demand is shifting towards higher, more standardized levels of quality, safety, and traceability [[3](https:\u002F\u002Fwww.mdpi.com\u002F2304-8158\u002F15\u002F16\u002F2864#B3-foods-15-02864)]. On the other hand, structural issues within the industry are intensifying. Major tea-producing regions worldwide are experiencing an aging rural population and a migration of young labor to urban areas, making it increasingly difficult to secure sufficient labor for harvest and other tasks, while costs continue to rise [[4](https:\u002F\u002Fwww.mdpi.com\u002F2304-8158\u002F15\u002F16\u002F2864#B4-foods-15-02864)]. Furthermore, traditional production is heavily reliant on empirical judgment from fertilization and pest control to harvesting standards and processing, hindering standardization, resulting in inconsistent product quality and difficulties in achieving scalable, high-grade output [[5](https:\u002F\u002Fwww.mdpi.com\u002F2304-8158\u002F15\u002F16\u002F2864#B5-foods-15-02864)]. This reliance on subjective, experience-based methods represents a fundamental bottleneck to modernization and sustainable development [[6](https:\u002F\u002Fwww.mdpi.com\u002F2304-8158\u002F15\u002F16\u002F2864#B6-foods-15-02864)].\n\nThe advent of new-generation AI, particularly deep learning, has driven transformative breakthroughs across various fields, including agriculture [[7](https:\u002F\u002Fwww.mdpi.com\u002F2304-8158\u002F15\u002F16\u002F2864#B7-foods-15-02864)]. Within this, AI-powered CV stands out as a key to enabling technology. Capable of automated recognition, measurement, and analysis from image and video data, CV can match or surpass human visual perception. This offers a powerful toolkit to address the traditional constraints of the tea industry [[8](https:\u002F\u002Fwww.mdpi.com\u002F2304-8158\u002F15\u002F16\u002F2864#B8-foods-15-02864)]. Applications range from macro-scale monitoring of plantation growth and nutrient status using drone-mounted multispectral cameras [[9](https:\u002F\u002Fwww.mdpi.com\u002F2304-8158\u002F15\u002F16\u002F2864#B9-foods-15-02864),[10](https:\u002F\u002Fwww.mdpi.com\u002F2304-8158\u002F15\u002F16\u002F2864#B10-foods-15-02864)] to real-time pest and disease identification via CNNs and precise detection of harvestable tea buds in complex natural environments using object detection algorithms [[11](https:\u002F\u002Fwww.mdpi.com\u002F2304-8158\u002F15\u002F16\u002F2864#B11-foods-15-02864)]. Moreover, technologies like hyperspectral imaging (HSI) enable non-destructive, real-time monitoring of biochemical constituents during critical processing stages such as withering and fermentation, facilitating digital and precise process control [[12](https:\u002F\u002Fwww.mdpi.com\u002F2304-8158\u002F15\u002F16\u002F2864#B12-foods-15-02864)].\n\nTo provide a comprehensive yet focused overview, this review adopts a pragmatic definition of CV that extends beyond conventional image-based techniques. Specifically, we include not only RGB, HSI, and thermal imaging (TIC) but also near-infrared spectroscopy (NIRS), LiDAR, and UAV [[13](https:\u002F\u002Fwww.mdpi.com\u002F2304-8158\u002F15\u002F16\u002F2864#B13-foods-15-02864)]. These technologies can be categorized into imaging modalities (RGB, HSI, TIC), spectroscopic techniques (NIRS), 3D spatial sensors (LiDAR), and aerial platforms (UAV). NIRS is considered because, in tea processing, it is often integrated with imaging systems for real-time quality prediction [[14](https:\u002F\u002Fwww.mdpi.com\u002F2304-8158\u002F15\u002F16\u002F2864#B14-foods-15-02864)]; LiDAR is included for its ability to provide three-dimensional spatial information that complements 2D images for bud localization and canopy analysis [[15](https:\u002F\u002Fwww.mdpi.com\u002F2304-8158\u002F15\u002F16\u002F2864#B15-foods-15-02864)], and UAV is treated as a platform that carries vision sensors, enabling large-scale, multi-temporal monitoring [[11](https:\u002F\u002Fwww.mdpi.com\u002F2304-8158\u002F15\u002F16\u002F2864#B11-foods-15-02864)]. The selection of these technologies was guided by three practical criteria-scalability (RGB cameras and UAVs can cover large areas), robustness under field conditions (LiDAR’s insensitivity to lighting variation), and feasibility for real-world deployment (NIRS portability for inline processing). This broad yet practical scope reflects the multimodal sensing strategies commonly employed in current tea industry research. Given this focus on practical application, we prioritize studies that report quantitative performance metrics, whether conducted in the laboratory, pilot line, or field.\n\nThe literature covered in this review was primarily identified through searches in Web of Science, Scopus, and Google Scholar using keywords including “computer vision,” “deep learning,” “tea industry,” “precision agriculture,” “hyperspectral imaging,” “NIRS,” “LiDAR,” and “UAV-based remote sensing.” The search covered publications from 2018 to 2025, with emphasis on studies reporting quantitative performance metrics. References were selected based on their relevance to the tea production chain and their contribution to the comparative analysis of vision technologies and algorithms.\n\nRecent reviews have summarized AI applications in tea management, harvesting, and processing, but a quantitative, full-chain comparison of vision technologies with a dedicated algorithm selection framework has been less explored. Unlike existing reviews that focus on isolated tasks or general agriculture, this review provides a quantitative comparison of vision technologies alongside a task-oriented algorithm selection framework across the entire tea production chain. It offers critical comparative analyses of technological approaches, identifies practical constraints, and examines translational challenges from lab to field, aiming to bridge the gap between research and industrial adoption.\n\n## 2. Key Enabling Technologies and Data Acquisition Systems in the Tea Industry\n\n### 2.1. AI-Powered CV Technology: Core Concepts for Tea Applications\n\nAI-powered CV integrates CV and deep learning to automatically extract information from images. Its pipeline typically includes data acquisition using RGB, HSI, TIC, or LiDAR sensors, data processing (preprocessing, segmentation), feature learning (via CNNs, autoencoders), and task execution (classification, detection, segmentation). In the tea industry, deep learning, especially CNNs, has become dominant for handling complex, unstructured scenes such as pest detection and bud recognition, while traditional ML (SVM, PLSR) remains useful for small-sample or near-linear problems. The operational framework of a CV system typically mirrors human visual processing, encompassing four key stages: data acquisition, data processing, data learning, and task execution (see [Figure 1](https:\u002F\u002Fwww.mdpi.com\u002F2304-8158\u002F15\u002F16\u002F2864#fig_body_display_foods-15-02864-f001)).\n\n**Figure 1.** Typical workflow of CV systems for tea applications. The red circles and lines indicate the propagation paths selected by the proposed strategy for the convolutional neural network.\n\n### 2.2. Multidimensional Visual Perception Technology\n\nThe efficacy of CV systems in the tea industry is fundamentally dependent on the underlying perception technologies for information acquisition. The detailed comparison of these technologies, including their core benefits, key applications, performance examples, validation settings, cost, maturity levels, and spatial resolution, is presented in [Table 1](https:\u002F\u002Fwww.mdpi.com\u002F2304-8158\u002F15\u002F16\u002F2864#table_body_display_foods-15-02864-t001). Unlike human vision, which is confined to the visible spectrum, modern perception technologies construct a multidimensional visual perception system that vastly exceeds human sensory capabilities by expanding the spectral range and integrating spatial, temporal, and chemical dimensions. This system provides holistic insights from macro to micro scales and from external morphology to internal chemistry, enabling intelligent monitoring and decision-making across the entire tea production chain, from cultivation to processing.\n\n**Table 1.** Comparison of multidimensional vision technologies in the tea industry.\n\nIn terms of spatial resolution (SR) and acquisition speed, the six technology categories differ substantially. RGB cameras typically achieve 0.1–1 mm\u002Fpixel at 30–60 fps. HSI offers similar spatial resolution but much slower acquisition (0.1–10 fps) due to scanning. NIRS provides no spatial image (point\u002Fline scan) but is fast (10–100 spectra\u002Fs). TIC gives lower spatial resolution (1–5 mm\u002Fpixel) at 1–30 fps. LiDAR generates 3D point clouds with cm-to-dm resolution at 10–100 k points\u002Fs. UAV-based remote sensing covers large areas at 0.5–5 cm\u002Fpixel per flight (10–30 min per 50 ha). A rough cost ranking (low to high) is: RGB \u003C NIRS \u003C TIC ≈ UAV \u003C HSI ≈ LiDAR. At the lower end, RGB cameras and consumer-grade NIRS devices are relatively affordable and widely available. TIC and UAV systems fall into the medium range, though costs vary considerably depending on sensor resolution and payload capacity. HSI and LiDAR remain the most expensive, due to their high-precision optics, complex calibration procedures, and substantial data storage and computational demands. To facilitate the comparison of technological maturity, each technology is assessed using the Technology Readiness Level (TRL), where levels 1–3 indicate basic research and proof-of-concept, 4–6 indicate validation in laboratory or relevant environments, and 7–9 indicate system prototype demonstration and full operational deployment.\n\n#### 2.2.1. RGB Imaging\n\nRGB imaging, which simulates human trichromatic vision, is the most widely adopted visual perception technology due to its low cost and simple deployment. In the tea industry, RGB cameras are primarily used to capture morphological, color, and textural features for tasks such as identifying bud maturity, preliminary screening of pronounced foliar disease lesions, and grading the appearance and color of dry tea. However, RGB imaging is inherently limited to surface reflectance in the visible spectrum. It cannot penetrate surfaces, detect chemical constituents, or effectively identify early-stage, latent physiological changes. Its performance is also highly susceptible to variations in ambient lighting conditions [[13](https:\u002F\u002Fwww.mdpi.com\u002F2304-8158\u002F15\u002F16\u002F2864#B13-foods-15-02864)]. To mitigate these lighting-induced issues, common preprocessing strategies include histogram equalization, white balance correction, and conversion to HSV or Lab color spaces to separate luminance from chrominance [[25](https:\u002F\u002Fwww.mdpi.com\u002F2304-8158\u002F15\u002F16\u002F2864#B25-foods-15-02864)]. Adaptive thresholding and illumination normalization can further improve robustness in non-uniform lighting [[26](https:\u002F\u002Fwww.mdpi.com\u002F2304-8158\u002F15\u002F16\u002F2864#B26-foods-15-02864)].\n\n#### 2.2.2. HSI\n\nHSI, which integrates spectroscopy with imaging, is a core enabling technology for intelligent perception in the tea industry [[27](https:\u002F\u002Fwww.mdpi.com\u002F2304-8158\u002F15\u002F16\u002F2864#B27-foods-15-02864)]. Unlike an RGB pixel containing only three broad bands, a hyperspectral pixel captures reflectance across hundreds of contiguous narrow bands (typically with spectral resolution \u003C 10 nm), forming a rich data cube with combined spatial and spectral dimensions [[28](https:\u002F\u002Fwww.mdpi.com\u002F2304-8158\u002F15\u002F16\u002F2864#B28-foods-15-02864)].\n\nThis capability stems from the absorption of specific near-infrared photons by vibrational overtones of chemical bonds involving hydrogen (O-H, N-H, C-H). Consequently, subtle changes in key internal chemical components produce distinctive responses in the spectral signature [[18](https:\u002F\u002Fwww.mdpi.com\u002F2304-8158\u002F15\u002F16\u002F2864#B18-foods-15-02864)]. HSI thus enables non-destructive, spatially resolved chemical mapping—for instance, visualizing moisture distribution during withering, quantifying the dynamics of polyphenol transformation during fermentation, or identifying pre-visual plant stress [[19](https:\u002F\u002Fwww.mdpi.com\u002F2304-8158\u002F15\u002F16\u002F2864#B19-foods-15-02864)]. In contrast, multispectral imaging (MSI) acquires data at only a few discrete bands, trading spectral resolution for lower cost and faster acquisition. While HSI typically provides higher prediction accuracy for internal chemical constituents (R 2> 0.91 for nitrogen and polyphenols [[19](https:\u002F\u002Fwww.mdpi.com\u002F2304-8158\u002F15\u002F16\u002F2864#B19-foods-15-02864)]), MSI is more suitable for large-scale, real-time field monitoring due to its reduced data volume [[29](https:\u002F\u002Fwww.mdpi.com\u002F2304-8158\u002F15\u002F16\u002F2864#B29-foods-15-02864)].\n\nThe primary challenges associated with HSI are data redundancy and high computational cost [[30](https:\u002F\u002Fwww.mdpi.com\u002F2304-8158\u002F15\u002F16\u002F2864#B30-foods-15-02864)]. Effective implementation often requires dimensionality reduction via wavelength selection algorithms or leveraging deep learning models like CNNs to automatically extract critical spectral-spatial features [[31](https:\u002F\u002Fwww.mdpi.com\u002F2304-8158\u002F15\u002F16\u002F2864#B31-foods-15-02864)].\n\n#### 2.2.3. NIRS\n\nNIRS operates on a similar spectroscopic principle to HSI but typically acquires spectral data in a point or line-scan mode, emphasizing rapid, portable, and online analysis [[6](https:\u002F\u002Fwww.mdpi.com\u002F2304-8158\u002F15\u002F16\u002F2864#B6-foods-15-02864)]. Its high sensitivity to hydrogen-containing groups makes it indispensable for real-time quality control during tea processing [[20](https:\u002F\u002Fwww.mdpi.com\u002F2304-8158\u002F15\u002F16\u002F2864#B20-foods-15-02864)]. The miniaturization of NIRS with fiber-optic probes facilitates direct integration into production lines for non-destructive, inline monitoring. Key applications include real-time moisture monitoring during fixation and drying for precise thermal control, and objective assessment of fermentation degree by tracking specific spectral features [[32](https:\u002F\u002Fwww.mdpi.com\u002F2304-8158\u002F15\u002F16\u002F2864#B32-foods-15-02864)].\n\nA current research frontier involves fusing NIR spectral data with RGB image data to leverage both chemical and morphological information for building more robust quality prediction models.\n\n#### 2.2.4. TIC and LiDAR\n\nTIC captures temperature distribution by detecting an object’s emitted infrared radiation. In tea garden management, it can identify canopy temperature anomalies indicative of water stress, providing a direct guide for precision irrigation [[33](https:\u002F\u002Fwww.mdpi.com\u002F2304-8158\u002F15\u002F16\u002F2864#B33-foods-15-02864)]. However, TIC accuracy is highly sensitive to environmental conditions (ambient temperature, wind, humidity, solar radiation) and leaf emissivity (typically 0.95–0.98 for green vegetation). To obtain reliable data, calibration is essential; common practices include using a reference blackbody, correcting for reflected temperature, applying emissivity adjustment, and measuring under stable weather conditions (midday, clear sky). Without such steps, temperature errors can exceed 2–3 °C, leading to incorrect stress diagnosis [[21](https:\u002F\u002Fwww.mdpi.com\u002F2304-8158\u002F15\u002F16\u002F2864#B21-foods-15-02864)].\n\nLiDAR, in contrast, provides precise depth information to generate 3D point clouds. This technology is critical for intelligent harvesting, enabling accurate 3D localization of tea buds within complex, overlapping foliage and guiding robotic arms for obstacle avoidance and grasping [[34](https:\u002F\u002Fwww.mdpi.com\u002F2304-8158\u002F15\u002F16\u002F2864#B34-foods-15-02864)]. Furthermore, LiDAR mounted on UAVs can create detailed 3D digital maps of tea gardens for canopy structure analysis and biomass estimation. However, dense canopy occlusion limits LiDAR performance: laser pulses are often intercepted by upper leaves, resulting in incomplete point clouds of lower buds. Multi-return or full-waveform LiDAR partially mitigates this but still struggles with deep internal structures [[22](https:\u002F\u002Fwww.mdpi.com\u002F2304-8158\u002F15\u002F16\u002F2864#B22-foods-15-02864)].\n\n#### 2.2.5. UAV\n\nUnmanned Aerial Systems equipped with multispectral, HSI, or RGB cameras form an aerial platform for macro-scale monitoring [[24](https:\u002F\u002Fwww.mdpi.com\u002F2304-8158\u002F15\u002F16\u002F2864#B24-foods-15-02864)]. They enable rapid, large-area acquisition of high-resolution imagery over difficult terrain. These systems are used to monitor field vigor and uniformity, derive vegetation indices like Leaf Area Index, assess nutrient status, and detect early-stage pest or disease outbreak foci [[23](https:\u002F\u002Fwww.mdpi.com\u002F2304-8158\u002F15\u002F16\u002F2864#B23-foods-15-02864)]. By analyzing multi-temporal r","MDPI Foods 2026-08-17","2026-08-17T00:00:00Z","论文",10,false,84,{"impact":17,"substance":18,"depth":19,"authority":20,"freshness":13,"relevant":21,"comment":22},18,22,20,14,1,"系统综述比较多种视觉技术在茶叶全产业链的应用，提供算法选择框架和具体性能数据，对智慧农业细分领域有实质参考价值。",[24],{"name":10,"url":6},[26,27,28,29],"综述","计算机视觉","深度学习","茶叶",{"code":31,"slug":32,"name":33,"path":34,"parent_code":35},"I2","agricultural-ai","农业AI与大模型","\u002Ftopic\u002Fdigital-agriculture\u002Fagricultural-ai","I",[],{"status":38,"locked":14,"version":39,"method":40,"evidence":41},"classified","2026-10-22.2","rules",{"candidates":42,"agricultural_evidence":45},[43,52],{"code":31,"score":44,"supported":45,"title":46,"abstract":48,"tags":51},19,true,[47],"computer vision",[28,49,27,50],"机器学习","YOLO",[28,27],{"code":53,"score":54,"supported":45,"title":55,"abstract":56,"tags":57},"B3",4,[],[29],[29],[59,60],"计算机视觉 深度学习 茶叶","计算机视觉 深度学习","计算机视觉深度学习茶叶-677",3,"2026-08-17T23:54:56.068084Z",{"total":65,"page":21,"page_size":65,"items":66},6,[67,132,204,260,312,374],{"id":68,"title":69,"url":70,"summary":71,"summary_zh":72,"content":8,"source_name":73,"source_url":70,"published_at":74,"category":12,"cover_url":8,"hotness":13,"is_selected":14,"score":75,"score_detail":76,"sources":80,"tags":82,"primary_topic":86,"secondary_topics":87,"classification":88,"search_phrases":114,"slug":117,"view_count":118,"doi":119,"paper":120,"created_at":131},5385,"A Comprehensive Review of Deep Learning and Computer Vision Techniques for Fruit Quality Assessment and Classification","https:\u002F\u002Fdoi.org\u002F10.25258\u002Fijddt.16.75s.151","Fruit image analysis has become an essential research area in precision agriculture, enabling automated fruit classification, quality grading, maturity estimation, disease diagnosis, yield prediction, and intelligent harvesting through advances in computer vision and artificial intelligence (AI).Recent progress in imaging technologies, large-scale benchmark datasets, and deep learning has substantially improved the accuracy, robustness, and scalability of automated fruit analysis systems.This survey presents a comprehensive and systematic review of image processing and AI techniques for fruit image analysis.It covers the complete image analysis pipeline, including image acquisition, preprocessing, enhancement, color space transformation, segmentation, feature extraction, and handcrafted texture descriptors.Furthermore, conventional machine learning algorithms are reviewed alongside state-of-the-art deep learning architectures, including Convolutional Neural Networks (CNNs), Vision Transformers (ViTs), and Vision Mamba models, for applications such as fruit recognition, quality assessment, disease detection, ripeness estimation, and object detection.Publicly available benchmark datasets, evaluation metrics, and performance comparison strategies are also summarized to facilitate objective assessment of existing approaches.In addition, the survey discusses key research challenges, including illumination variations, occlusion, complex orchard environments, limited annotated datasets, computational complexity, and domain adaptation, while highlighting emerging research trends such as self-supervised learning, multimodal learning, explainable AI, lightweight deep networks, foundation models, and edge intelligence for real-time deployment.By providing a unified review of methodologies, datasets, evaluation protocols, challenges, and future research directions, this survey offers a comprehensive reference for researchers and practitioners developing intelligent and scalable fruit image analysis systems for precision agriculture.","水果图像分析已成为精准农业中的一个重要研究领域，借助计算机视觉和人工智能（AI）的进步，实现了果实自动分类、品质分级、成熟度估计、病害诊断、产量预测和智能采收。成像技术、大规模基准数据集和深度学习的最新进展显著提升了自动化水果分析系统的准确性、鲁棒性和可扩展性。本综述对用于水果图像分析的图像处理和AI技术进行了全面而系统的回顾。内容涵盖完整的图像分析流程，包括图像采集、预处理、增强、颜色空间变换、分割、特征提取和手工纹理描述符。此外，综述了传统机器学习算法以及最先进的深度学习架构，包括卷积神经网络（CNN）、视觉Transformer（ViT）和视觉Mamba模型，用于果实识别、品质评估、病害检测、成熟度估计和目标检测等应用。同时总结了公开可用的基准数据集、评价指标和性能比较策略，以便对现有方法进行客观评估。此外，本综述讨论了关键研究挑战，包括光照变化、遮挡、复杂果园环境、标注数据集有限、计算复杂度和域适应，并重点介绍了新兴研究趋势，如自监督学习、多模态学习、可解释AI、轻量级深度网络、基础模型和面向实时部署的边缘智能。通过提供方法、数据集、评价协议、挑战和未来研究方向的统一综述，本综述为开发面向精准农业的智能且可扩展的水果图像分析系统的研究人员和实践者提供了全面的参考。","International Journal of Drug Delivery Technology","2026-10-18T00:00:00Z",56,{"impact":77,"substance":17,"depth":78,"authority":65,"freshness":77,"relevant":21,"comment":79},8,16,"系统性综述覆盖CNN、ViT、Vision Mamba等前沿方法，信息量足，但期刊权威性一般且偏学术综述，公共价值有限，适合主题聚合而非每日精选头条。",[81],{"name":73,"url":70},[83,84,27,28,85],"智慧农业","农业人工智能","水果品质检测",{"code":31,"slug":32,"name":33,"path":34,"parent_code":35},[],{"status":38,"locked":14,"version":39,"method":40,"evidence":89},{"candidates":90,"agricultural_evidence":45},[91,100,108],{"code":31,"score":92,"supported":45,"title":93,"abstract":95,"tags":99},28,[94,47],"deep learning",[96,28,49,27,97,94,98,47],"目标检测","machine learning","artificial intelligence",[84,28,27],{"code":101,"score":102,"supported":45,"title":103,"abstract":104,"tags":107},"I1",7,[],[105,106],"精准农业","precision agriculture",[83],{"code":109,"score":62,"supported":45,"title":110,"abstract":111,"tags":113},"B2",[],[112],"orchard",[],[115,116],"水果品质 计算机视觉 深度学习","农业人工智能 水果品质检测 计算机视觉 智慧农业","水果品质计算机视觉深度学习-5385",0,"10.25258\u002Fijddt.16.75s.151",{"doi":119,"openalex_id":121,"authors":122,"venue":73,"cited_by_count":118,"oa_url":70,"card":123,"direction":129,"ingested_from":130},"W7222892831",[],{"tldr":124,"method":125,"finding":126,"direction":127,"opportunity":128},"系统综述深度学习与计算机视觉在水果品质评估与分级中的全流程方法。","综述图像采集到分割、特征提取及CNN、ViT、Vision Mamba等模型与公","深度学习显著提升水果分析精度与鲁棒性，但光照、遮挡和小样本仍是主要瓶颈。","农业遥感与作物表型","可探索自监督与轻量化模型在复杂果园遮挡、小样本场景下的边缘实时部署。","智慧农业 \u002F 农业物联网","openalex","2026-10-20T23:30:10.066042Z",{"id":133,"title":134,"url":135,"summary":136,"summary_zh":137,"content":8,"source_name":138,"source_url":135,"published_at":139,"category":12,"cover_url":8,"hotness":13,"is_selected":14,"score":140,"score_detail":141,"sources":145,"tags":147,"primary_topic":149,"secondary_topics":154,"classification":156,"search_phrases":173,"slug":176,"view_count":118,"doi":177,"paper":178,"created_at":203},4744,"An empirical performance comparison of deep learning models for apple leaf disease recognition","https:\u002F\u002Fdoi.org\u002F10.1007\u002Fs10791-026-10573-4","Currently, Biotic and abiotic factors, in agriculture, are estimated to be responsible for around 22% of the total losses in crop production. It is therefore very important, but at the same time very difficult for farmers to identify these constraints at the earliest possible stage. The most recent developments in computer vision have made it possible to identify as well as classify plant lesions at a primary level. Various deep learning techniques have made it possible to detect biotic stress in apple plant leaves with a very high degree of accuracy. Hence, automated segmentation of images of diseased leaves has become a leading research theme all over the world. In this paper, a dataset from FGVC7, which is an open, source dataset and contains four different classes of leaf diseases of the apple crop is used to measure the performance of the proposed model. A three, step image preprocessing procedure is conducted, which involves image enhancement, removal of background noise, and object identification through a variety of approaches. Using UNet, PSPNet, and UNet++, disease areas are identified. UNet + + delivering the most precise object segmentation. The segmented patch data are then fed into a fusion, based classification model as input. Patch segmentation makes it possible to separate the various disease areas, thus increasing the classification accuracy. The proposed framework combines several CNN architectures, namely, VGG16, InceptionV3, AlexNet, and GoogleNet, in order to take advantage of their complementary strengths, then a Vision Transformer (ViT) is used to identify distinct local features via early feature fusion. After this, the efficiency of the proposed model is check in comparison with the pre-trained CNN models individually. Comparing with all other models the proposed model has shown the accuracy of highest order i.e. 98.85%, clearly out performing individual CNN models.","目前，农业中的生物和非生物因素估计约占作物生产总损失的22%。因此，农民尽早识别这些限制因素非常重要，但同时也非常困难。计算机视觉的最新发展使得在初级水平上识别和分类植物病变成为可能。各种深度学习技术使得以极高准确度检测苹果植株叶片的生物胁迫成为可能。因此，患病叶片图像的自动分割已成为世界范围内的前沿研究主题。本文使用来自FGVC7的数据集来衡量所提出模型的性能，该数据集是一个开放源数据集，包含苹果作物的四类不同叶片病害。进行了三步图像预处理，包括图像增强、背景噪声去除以及通过多种方法进行目标识别。使用UNet、PSPNet和UNet++识别病害区域，其中UNet++提供了最精确的目标分割。然后将分割后的图像块数据作为输入馈入基于融合的分类模型。图像块分割使得能够分离不同的病害区域，从而提高分类准确率。所提出的框架结合了多种CNN架构，即VGG16、InceptionV3、AlexNet和GoogleNet，以利用它们互补的优势，然后使用视觉Transformer（ViT）通过早期特征融合来识别不同的局部特征。此后，将所提出模型的效率与各个预训练CNN模型进行比较。与所有其他模型相比，所提出的模型显示出最高阶的准确率，即98.85%，明显优于各个单独的CNN模型。","Discover Computing","2026-10-10T00:00:00Z",69,{"impact":142,"substance":19,"depth":78,"authority":143,"freshness":77,"relevant":21,"comment":144},12,13,"方法组合（UNet++分割+CNN与ViT融合）与98.85%准确率有参考价值，但属实验室数据集验证，产业落地影响有限，可作为技术前沿资讯收录。",[146],{"name":138,"url":135},[83,84,27,28,148],"苹果病害识别",{"code":109,"slug":150,"name":151,"path":152,"parent_code":153},"fruit","果树","\u002Ftopic\u002Fhorticulture\u002Ffruit","B",[155],{"code":31,"slug":32,"name":33,"path":34,"parent_code":35},{"status":38,"locked":14,"version":39,"method":40,"evidence":157},{"candidates":158,"agricultural_evidence":45},[159,163,169],{"code":31,"score":19,"supported":45,"title":160,"abstract":161,"tags":162},[94],[28,27,94,47],[84,28,27],{"code":109,"score":20,"supported":45,"title":164,"abstract":166,"tags":168},[165],"apple",[167,165],"苹果",[],{"code":101,"score":21,"supported":14,"title":170,"abstract":171,"tags":172},[],[],[83],[174,175],"苹果叶片病害 深度学习","FGVC7 苹果病害 数据集","苹果叶片病害深度学习-4744","10.1007\u002Fs10791-026-10573-4",{"doi":177,"openalex_id":179,"authors":180,"venue":138,"cited_by_count":118,"oa_url":196,"card":197,"direction":129,"ingested_from":130},"W7221353656",[181,183,186,188,190,193],{"name":182,"orcid":8},"Apoorva Arora",{"name":184,"orcid":185},"Vinay Gautam","https:\u002F\u002Forcid.org\u002F0000-0002-0258-5132",{"name":187,"orcid":8},"Anand Muni Mishra",{"name":189,"orcid":8},"Santar Pal Singh",{"name":191,"orcid":192},"Yadvendra Pratap Singh","https:\u002F\u002Forcid.org\u002F0000-0001-6396-3355",{"name":194,"orcid":195},"Mukund Pratap Singh","https:\u002F\u002Forcid.org\u002F0000-0001-5877-9510","https:\u002F\u002Flink.springer.com\u002Fcontent\u002Fpdf\u002F10.1007\u002Fs10791-026-10573-4.pdf",{"tldr":198,"method":199,"finding":200,"direction":201,"opportunity":202},"用UNet++分割苹果病叶并融合多种CNN与ViT分类，实现98.85%识别准确率。","FGVC7数据集，UNet\u002FPSPNet\u002FUNet++分割，VGG16等CNN与","UNet++分割最准，融合模型准确率达98.85%，优于单个CNN模型。","农业人工智能与决策模型","可探索轻量化分割-分类联合模型在田间移动端的实时部署与多作物泛化。","2026-10-12T23:30:09.091475Z",{"id":205,"title":206,"url":207,"summary":208,"summary_zh":209,"content":8,"source_name":210,"source_url":207,"published_at":211,"category":12,"cover_url":8,"hotness":13,"is_selected":14,"score":212,"score_detail":213,"sources":216,"tags":218,"primary_topic":221,"secondary_topics":222,"classification":223,"search_phrases":240,"slug":243,"view_count":118,"doi":244,"paper":245,"created_at":259},5587,"Calibration-Shift Inference for Computer Vision Agriculture Yield Models Under Retrofit Sensor Replacement: An Artificial Intelligence Post-Deployment Recalibration Guarantee","https:\u002F\u002Fdoi.org\u002F10.5281\u002Fzenodo.23544195","Computer vision yield models deployed in agricultural fields are routinely paired with retrofit sensor packages that replace or supplement the original imaging hardware, and the resulting calibration shift can silently invalidate the model's predictive guarantees. The analysis characterises the operating characteristics of the design under a declared variance structure with between-cluster standard deviation 6.0, residual standard deviation 4.5 and intraclass correlation 0.64, using 2000 replications per design cell across cluster counts of 8, 16, 24 and 32 and effect sizes of 3, 5 and 8 percentage points. The design attains 80% power at a 5 percentage point effect with 54 clusters. The paper states a formal guarantee: under stated assumptions on the stability of the calibration map, the before-after-control-impact estimator yields a consistent and asymptotically normal estimate of the calibration shift, and the associated confidence interval covers the true shift at the nominal rate. The contribution is a design-theoretic account of how retrofit-induced calibration shift can be detected and bounded without observational data, together with the sample-size requirements that follow.","部署在农田中的计算机视觉产量模型通常与改装传感器组件配合使用，这些组件用于替换或补充原始成像硬件，由此产生的校准偏移可能悄然使模型的预测保证失效。该分析在声明的方差结构下刻画了设计的工作特性，其中组间标准差为6.0、残差标准差为4.5、组内相关系数为0.64，每个设计单元采用2000次重复，聚类数分别为8、16、24和32，效应量为3、5和8个百分点。该设计在54个聚类、5个百分点效应下达到80%的检验力。论文给出了一项形式化保证：在关于校准映射稳定性的既定假设下，前后对照影响（before-after-control-impact）估计量给出校准偏移的一致且渐近正态的估计，且相关置信区间以名义覆盖率覆盖真实偏移。其贡献在于从设计理论角度说明了如何在没有观测数据的情况下检测并界定改装引起的校准偏移，以及随之而来的样本量要求。","Zenodo (CERN European Organization for Nuclear Research)","2026-10-21T00:00:00Z",65,{"impact":77,"substance":17,"depth":214,"authority":142,"freshness":13,"relevant":21,"comment":215},17,"方法学扎实、统计设计严谨的预印本论文，但偏理论设计层面，缺乏田间实证数据，产业影响有限，适合作为技术参考而非每日精选头条。",[217],{"name":210,"url":207},[83,84,27,219,220],"产量预测","传感器校准",{"code":31,"slug":32,"name":33,"path":34,"parent_code":35},[],{"status":38,"locked":14,"version":39,"method":40,"evidence":224},{"candidates":225,"agricultural_evidence":45},[226,231,235],{"code":31,"score":227,"supported":45,"title":228,"abstract":229,"tags":230},24,[98,47],[27,47],[84,27],{"code":101,"score":21,"supported":14,"title":232,"abstract":233,"tags":234},[],[],[83],{"code":236,"score":21,"supported":14,"title":237,"abstract":238,"tags":239},"I3",[],[],[219],[241,242],"计算机视觉 产量预测 传感器校准","农业人工智能 传感器校准 计算机视觉 产量预测","计算机视觉产量预测传感器校准-5587","10.5281\u002Fzenodo.23544195",{"doi":244,"openalex_id":246,"authors":247,"venue":210,"cited_by_count":118,"oa_url":207,"card":254,"direction":201,"ingested_from":130},"W7223104380",[248,250,252],{"name":249,"orcid":8},"Tigist Alemu Gebremariam",{"name":251,"orcid":8},"Hawi Tesfaye Desta",{"name":253,"orcid":8},"Mulugeta Bekele Wolde",{"tldr":255,"method":256,"finding":257,"direction":201,"opportunity":258},"提出改造传感器替换后视觉产量模型校准漂移的检测与置信区间保证方法。","用前后对照影响估计量，基于方差结构做2000次重复的模拟设计分析。","5个百分点效应下54个集群可达80%检验力，估计量一致且渐近正态。","可将该设计理论扩展到多传感器异构替换与在线校准监测，填补真实田间验证空白。","2026-10-22T11:21:01.554995Z",{"id":261,"title":262,"url":263,"summary":264,"summary_zh":265,"content":8,"source_name":210,"source_url":263,"published_at":211,"category":12,"cover_url":8,"hotness":13,"is_selected":14,"score":266,"score_detail":267,"sources":269,"tags":271,"primary_topic":274,"secondary_topics":275,"classification":276,"search_phrases":292,"slug":295,"view_count":118,"doi":296,"paper":297,"created_at":311},5584,"Distributionally-Robust Evaluation of Computer Vision Agriculture Disease Classifiers Under Climate-Driven Pathogen Range Shifts: An Artificial Intelligence Worst-Case Generalization Bound","https:\u002F\u002Fdoi.org\u002F10.5281\u002Fzenodo.23544150","Computer vision classifiers for crop disease are evaluated almost exclusively on held-out samples drawn from the same distribution as their training data, yet climate-driven pathogen range shifts mean that deployment environments are systematically out of distribution. The framework treats the deployment distribution as an adversarial perturbation of the training distribution within a Wasserstein ball whose radius is calibrated to observed climate displacement. Using NASA POWER monthly reanalysis for Harare, Zimbabwe (2000-2023, 312 monthly observations), the analysis estimates a mean air temperature of 18.914 C, a standard deviation of 3.107 C, and a warming trend of 0.2001 C per decade. These estimates anchor the radius of the ambiguity set and the resulting bound. The bound decomposes worst-case risk into an empirical risk term, a Wasserstein radius term scaled by the Lipschitz constant of the classifier, and a concentration term. Under plausible Lipschitz constants for convolutional backbones, the radius term dominates the empirical risk term once the climate displacement exceeds a threshold that the Harare record already approaches. The contribution is a formal evaluation criterion that replaces the i. i. d. assumption with a climate-calibrated ambiguity set, together with the analytical machinery needed to compute it.","作物病害的计算机视觉分类器几乎完全在与训练数据同分布的留出样本上进行评估，然而气候驱动的病原体分布范围变化意味着部署环境系统性地处于分布之外。该框架将部署分布视为训练分布在Wasserstein球内的对抗性扰动，球的半径根据观测到的气候位移进行校准。利用NASA POWER对津巴布韦哈拉雷的月度再分析数据（2000—2023年，312个月度观测值），分析估计平均气温为18.914°C，标准差为3.107°C，每十年升温趋势为0.2001°C。这些估计值锚定了模糊集的半径及由此得到的界。该界将最坏情况风险分解为经验风险项、由分类器Lipschitz常数缩放后的Wasserstein半径项以及集中项。在卷积骨干网络合理的Lipschitz常数下，一旦气候位移超过哈拉雷记录已接近的阈值，半径项便主导经验风险项。本文的贡献在于提出了一种形式化评估准则，以气候校准的模糊集替代独立同分布（i.i.d.）假设，并提供了计算该准则所需的分析工具。",75,{"impact":142,"substance":18,"depth":17,"authority":143,"freshness":13,"relevant":21,"comment":268},"方法新颖、数据具体，但属方法论预印论文，产业影响有限，适合作为专业精选而非头条。",[270],{"name":210,"url":263},[83,84,27,272,273],"作物病害","气候变化",{"code":31,"slug":32,"name":33,"path":34,"parent_code":35},[],{"status":38,"locked":14,"version":39,"method":40,"evidence":277},{"candidates":278,"agricultural_evidence":45},[279,283,288],{"code":31,"score":227,"supported":45,"title":280,"abstract":281,"tags":282},[98,47],[27,47],[84,27],{"code":284,"score":21,"supported":14,"title":285,"abstract":286,"tags":287},"H7",[],[],[273],{"code":101,"score":21,"supported":14,"title":289,"abstract":290,"tags":291},[],[],[83],[293,294],"NASA POWER 哈拉雷 气温","Wasserstein 分布鲁棒 病害分类","NASAPOWER哈拉雷气温-5584","10.5281\u002Fzenodo.23544150",{"doi":296,"openalex_id":298,"authors":299,"venue":210,"cited_by_count":118,"oa_url":263,"card":306,"direction":201,"ingested_from":130},"W7223087892",[300,302,304],{"name":301,"orcid":8},"Kossi Afeli",{"name":303,"orcid":8},"Akouvi Amegashie",{"name":305,"orcid":8},"Komlan Sélom Dossou",{"tldr":307,"method":308,"finding":309,"direction":201,"opportunity":310},"提出气候校准的分布鲁棒评估准则，替代i.i.d.假设评价作物病害视觉分类器。","用Wasserstein球建模部署分布偏移，以NASA POWER哈拉雷气温数据","气候位移超阈值后，半径项主导经验风险，现有评估会高估模型泛化能力。","可将气候校准的鲁棒评估扩展到多作物多区域，并研究适应分布偏移的鲁棒训练方法。","2026-10-22T11:21:00.036937Z",{"id":313,"title":314,"url":315,"summary":316,"summary_zh":317,"content":8,"source_name":318,"source_url":315,"published_at":74,"category":12,"cover_url":8,"hotness":13,"is_selected":14,"score":319,"score_detail":320,"sources":322,"tags":324,"primary_topic":327,"secondary_topics":332,"classification":334,"search_phrases":354,"slug":357,"view_count":118,"doi":358,"paper":359,"created_at":373},5387,"LIGHTWEIGHT FEATURE FUSED ATTENTION DRIVEN MULTI LEVEL HYBRID MOBILENET V2 FRAMEWORK FOR EARLY TOMATO LEAF DISEASE DETECTION","https:\u002F\u002Fdoi.org\u002F10.62441\u002Factainnovations.vi.1918","It is crucial to find and treat tomato crop diseases early and accurately to improve crop efficiency and make sure there is no scarcity of crop. However, computerized analysis is provocative due to lighting differences, compound backgrounds, and graphic likenesses among infection regions. Research work proposed a lightweight Attention Driven Multi Level Hybrid MobileNet-V2 algorithm that fuses the handcrafted texture structures with Deep Learning (DL) features, then image preprocessing is incorporated using Gaussian, Bilateral, and Guided filters to minimize noise by preserving the disease lesions. LAB color model along with Contrast Limited Adaptive Histogram Equalization (CLAHE) are used to improve the contrast. Otsu thresholding is involved to effectively separate the leaf segment from diseased parts. Local Binary Pattern (LBP) and Gray Level Co-occurrence Matrix (GLCM) descriptors are used to obtain fine-grained features, while high level features are gained by pretrained MobileNet-V2 model. Features fusion is fed to the fully connected layers of deep learning model for training and validation purpose. Results have revealed that the suggested model attains better classification performance as compared to existing models, making it an effective explanation for real time tomato leaf disease detection for smart agriculture applications.","早期准确发现并治疗番茄作物病害，对于提高作物效率、确保作物不短缺至关重要。然而，由于光照差异、复杂背景以及感染区域之间的图形相似性，计算机化分析具有挑战性。研究提出了一种轻量级注意力驱动多层混合MobileNet-V2算法，该算法将手工纹理结构与深度学习（DL）特征相融合，然后结合高斯、双边和引导滤波器进行图像预处理，以在保留病害病变的同时最小化噪声。采用LAB颜色模型和对比度受限自适应直方图均衡化（CLAHE）来提高对比度。引入Otsu阈值分割以有效分离叶片片段与病害部分。使用局部二值模式（LBP）和灰度共生矩阵（GLCM）描述符获取细粒度特征，同时通过预训练的MobileNet-V2模型获取高层特征。特征融合后输入深度学习模型的全连接层进行训练和验证。结果表明，与现有模型相比，所提出的模型取得了更好的分类性能，使其成为智慧农业应用中实时番茄叶片病害检测的有效解决方案。","Acta Innovations",64,{"impact":142,"substance":17,"depth":78,"authority":13,"freshness":77,"relevant":21,"comment":321},"提出融合手工纹理特征与MobileNet-V2的轻量番茄叶病检测方法，有一定方法新颖性，但属单篇论文、应用范围有限，未达每日精选门槛。",[323],{"name":318,"url":315},[83,84,28,325,326],"番茄病害","图像识别",{"code":328,"slug":329,"name":330,"path":331,"parent_code":153},"B1","vegetables","蔬菜","\u002Ftopic\u002Fhorticulture\u002Fvegetables",[333],{"code":31,"slug":32,"name":33,"path":34,"parent_code":35},{"status":38,"locked":14,"version":39,"method":40,"evidence":335},{"candidates":336,"agricultural_evidence":45},[337,344,349],{"code":328,"score":338,"supported":45,"title":339,"abstract":341,"tags":343},15,[340],"tomato",[342,340],"番茄",[325],{"code":31,"score":345,"supported":45,"title":346,"abstract":347,"tags":348},9,[],[28,94],[84,326,28],{"code":101,"score":102,"supported":45,"title":350,"abstract":351,"tags":353},[],[83,352],"smart agriculture",[83],[355,356],"MobileNet-V2 番茄叶片病害","番茄叶病 深度学习 检测","MobileNet-V2番茄叶片病害-5387","10.62441\u002Factainnovations.vi.1918",{"doi":358,"openalex_id":360,"authors":361,"venue":318,"cited_by_count":118,"oa_url":367,"card":368,"direction":129,"ingested_from":130},"W7222977034",[362,364],{"name":363,"orcid":8},"Mrs. Vijetalaxmi A Hegde",{"name":365,"orcid":366},"Abhay Ramakant Kasetwar","https:\u002F\u002Forcid.org\u002F0000-0002-5800-8764","https:\u002F\u002Factainnovations.com\u002Findex.php\u002Fpub\u002Farticle\u002Fdownload\u002F1918\u002F981",{"tldr":369,"method":370,"finding":371,"direction":201,"opportunity":372},"提出轻量注意力多级混合MobileNet-V2框架，融合手工纹理与深度学习特征实现番茄叶片病害早期检","高斯\u002F双边\u002F引导滤波去噪，LAB+CLAHE增强，Otsu分割，LBP与GLCM","所提模型分类性能优于现有模型，可有效用于智慧农业实时番茄叶片病害检测。","可探索多作物跨域泛化、田间复杂光照下轻量模型部署与可解释性，及病害早期分级检测。","2026-10-20T23:30:10.424511Z",{"id":375,"title":376,"url":377,"summary":378,"summary_zh":8,"content":8,"source_name":379,"source_url":8,"published_at":380,"category":12,"cover_url":8,"hotness":13,"is_selected":14,"score":381,"score_detail":382,"sources":384,"tags":386,"primary_topic":389,"secondary_topics":395,"classification":397,"search_phrases":413,"slug":416,"view_count":118,"doi":8,"paper":417,"created_at":425},5363,"牛肉牛区域特征计算机视觉分析:基于 YOLO 与深度估计的牛体况自动评分","https:\u002F\u002Fbioengineer.org\u002Fcameras-learn-to-read-a-cows-spine-tail-head-and-body-shape-to-score-its-condition","Islam M.N., Schrick F.N., Edwards J.L., Burns R.T., Qi H., Gan H. (2026) 开发一种基于区域特定计算机视觉特征的牛肉牛体况自动评分(BCS)方法。研究通过 YOLO 检测 + 深度估计提取脊柱、尾根、体廓等区域特征,使用 SHAP 进行可解释 AI 分析,并以 XGBoost 融合。在 5 折交叉验证中两种模式均稳定。验证集仅 72 只动物且极端值样本少,深度图来自推断而非实测。框架可作为后续发展的方法学基础。DOI: 10.1016\u002Fj.atech.2026.102628。","Smart Agricultural Technology Vol.15 Article 102628 2026-10-15","2026-10-15T00:00:00Z",68,{"impact":142,"substance":19,"depth":214,"authority":143,"freshness":65,"relevant":21,"comment":383},"方法学上有新意，但验证样本仅72头且深度图为推断值，属细分领域论文进展，未达每日精选门槛。",[385],{"name":379,"url":377},[83,84,27,387,388],"肉牛养殖","体况评分",{"code":390,"slug":391,"name":392,"path":393,"parent_code":394},"C2","ruminants","草食牲畜","\u002Ftopic\u002Flivestock\u002Fruminants","C",[396],{"code":31,"slug":32,"name":33,"path":34,"parent_code":35},{"status":38,"locked":14,"version":39,"method":40,"evidence":398},{"candidates":399,"agricultural_evidence":45},[400,404,409],{"code":31,"score":227,"supported":45,"title":401,"abstract":402,"tags":403},[27,50],[27,50],[84,27],{"code":390,"score":142,"supported":45,"title":405,"abstract":407,"tags":408},[406],"肉牛",[406],[387],{"code":101,"score":21,"supported":14,"title":410,"abstract":411,"tags":412},[],[],[83],[414,415],"YOLO 深度估计 牛体况评分","牛肉牛 计算机视觉 BCS","YOLO深度估计牛体况评分-5363",{"doi":8,"openalex_id":8,"authors":418,"venue":8,"cited_by_count":118,"oa_url":8,"card":419,"direction":127,"ingested_from":424},[],{"tldr":420,"method":421,"finding":422,"direction":127,"opportunity":423},"用YOLO检测与深度估计提取牛体区域特征，实现肉牛体况自动评分。","YOLO检测、深度估计、SHAP可解释分析与XGBoost融合，5折交叉验证。","区域特征融合评分稳定，但验证集仅72头且深度图为推断值。","可引入实测深度传感器与更大样本，提升极端体况评分鲁棒性。","agent","2026-10-20T02:47:54.361149Z"]