A Lightweight Phenology-Aware YOLOv5 Framework for Tomato Growth Stage Detection in Resource-Constrained Bhutanese Greenhouse Environments

• URL: http://arxiv.org/abs/2608.30088v1
• Date: Sun, 30 Aug 2026 23:23:08 GMT
• Title: A Lightweight Phenology-Aware YOLOv5 Framework for Tomato Growth Stage Detection in Resource-Constrained Bhutanese Greenhouse Environments
• Abstract summary: In Bhutan, greenhouse cultivation is affected by altitude variability, large diurnal temperature fluctuations, diffuse illumination, limited automation, and a scarcity of locally annotated datasets.<n>This work proposes Pheno-Lite + Efficient Channel Attention (ECA), a lightweight, phenology-aware object detection architecture derived from Ultralytics YOLOv5 for tomato growth stage recognition.<n>The proposed model achieves 90.6% precision, 88.8% recall, and 92.6% mAP@50, with 4.0 million parameters and 10.9 GFLOPs at 640 x 640 resolution.
• Score: 0.3093890460224435
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Accurate detection of tomato growth stages is essential for stage-specific greenhouse management and precision agriculture. In Bhutan, greenhouse cultivation is affected by altitude variability, large diurnal temperature fluctuations, diffuse illumination, limited automation, and a scarcity of locally annotated datasets, limiting the applicability of conventional deep learning models. This work proposes Pheno-Lite + Efficient Channel Attention (ECA), a lightweight, phenology-aware object detection architecture derived from Ultralytics YOLOv5 for tomato growth stage recognition. A balanced dataset of 2,464 annotated images was constructed from locally collected greenhouse images in Bhutan and publicly available tomato images, with augmentation designed to simulate local greenhouse conditions. The dataset includes vegetative (820), flowering (824), fruiting (820), and background (26) samples. The proposed architecture introduces two customized backbone modules: C3 PhenoLite, which enhances spatial and texture feature extraction using depthwise residual refinement, and C3 ECA, which strengthens inter-channel feature interactions through efficient channel attention. The proposed model achieves 90.6% precision, 88.8% recall, and 92.6% mAP@50, with 4.0 million parameters and 10.9 GFLOPs at 640 x 640 resolution. These results demonstrate its potential for real-time and climate-resilient greenhouse deployment in Bhutan.

Related papers

• Visual-SLAM for the detection of hidden tomatoes in greenhouses by Hierarchical Localization and GLOMAP for robotized harvesting [39.55454029331185]
  This work proposes a Visual-SLAM system using a monocular camera, which is significantly more cost-effective and specifically tailored for agricultural applications.<n>Tests were carried out on a real tomato bunch, located in the Agroconnect experimental greenhouse.
  arXiv  Detail & Related papers  (2026-09-10T16:19:36Z)
• WeedExpert-R1: Incentivizing Botanical Reasoning in MLLMs with Reinforcement Learning for Precision Weed Grounding [5.155445137665098]
  This study introduces WeedExpert-R1, a multimodal model that learns visually grounded botanical reasoning through verifiable rewards.<n>WeedExpert-R1-4B achieved 75.82 percent exact-set precision at an IoU threshold of 0.5, 89.30 percent precision, and 87.81 percent recall.
  arXiv  Detail & Related papers  (2026-07-17T20:33:10Z)
• LeafInst - Unified Instance Segmentation Network for Fine-Grained Forestry Leaf Phenotype Analysis: A New UAV based Benchmark [10.61947524568352]
  LeafInst is a novel segmentation framework tailored for irregular and multi-scale leaf structures.<n>It achieves 68.4 mAP, outperforming YOLOv11 by 7.1 percent and MaskDINO by 6.5 percent.
  arXiv  Detail & Related papers  (2026-03-04T01:01:57Z)
• BAWSeg: A UAV Multispectral Benchmark for Barley Weed Segmentation [31.004130414489698]
  We propose a two-stream segmentation network that fuses radiance cues and normalized index cues at native resolution.<n>Vegetation-Index and Spectral Attention operates on vegetation-index maps with windowed self-attention.<n>Vegetation-Index and Spectral Attention achieves 75.6% mIoU and 63.5% weed IoU with 22.8M parameters, outperforming a multispectral SegFormer-B1 baseline by 1.2 mIoU and 1.9 weed IoU.
  arXiv  Detail & Related papers  (2026-03-02T14:49:05Z)
• Sapling-NeRF: Geo-Localised Sapling Reconstruction in Forests for Ecological Monitoring [36.7717316600672]
  Saps are key indicators of forest regeneration and overall forest health.<n>Traditional 3D sensing methods fail to capture fine-scale architectural traits.<n>We present a pipeline which fuses NeRF, LiDAR SLAM, andlings to enable repeatable, geo-localised monitoring of saplings.
  arXiv  Detail & Related papers  (2026-02-26T08:13:47Z)
• StomataSeg: Semi-Supervised Instance Segmentation for Sorghum Stomatal Components [27.879332209503435]
  Sorghum is a globally important cereal grown widely in water-limited and stress-prone regions.<n>Current methods still face challenges related to nested small structures and annotation bottlenecks.<n>We propose a semi-supervised instance segmentation framework tailored for analysis of sorghum stomatal components.
  arXiv  Detail & Related papers  (2026-01-31T13:01:02Z)
• A Domain-Adapted Lightweight Ensemble for Resource-Efficient Few-Shot Plant Disease Classification [0.0]
  We present a few-shot learning approach that combines domain-adapted MobileNetV2 and MobileNetV3 models as feature extractors.<n>For the classification task, the fused features are passed through a Bi-LSTM classifier enhanced with attention mechanisms.<n>It consistently improved performance across 1 to 15 shot scenarios, reaching 98.23+-0.33% at 15 shot.<n> Notably, it also outperformed the previous SOTA accuracy of 96.4% on six diseases from PlantVillage, achieving 99.72% with only 15-shot learning.
  arXiv  Detail & Related papers  (2025-12-15T15:17:29Z)
• Branched Broomrape Detection in Tomato Farms Using Satellite Imagery and Time-Series Analysis [0.2770822269241974]
  Branched broomrape (Phelipanche ramosa) threatens tomato production by extracting nutrients from the host.<n>We present an end-to-end pipeline that uses Sentinel-2 imagery and time-series analysis to identify broomrape-infested tomato fields in California.
  arXiv  Detail & Related papers  (2025-09-13T03:51:11Z)
• Data Augmentation and Resolution Enhancement using GANs and Diffusion Models for Tree Segmentation [49.13393683126712]
  Urban forests play a key role in enhancing environmental quality and supporting biodiversity in cities.<n> accurately detecting trees is challenging due to complex landscapes and the variability in image resolution caused by different satellite sensors or UAV flight altitudes.<n>We propose a novel pipeline that integrates domain adaptation with GANs and Diffusion models to enhance the quality of low-resolution aerial images.
  arXiv  Detail & Related papers  (2025-05-21T03:57:10Z)
• BonnBeetClouds3D: A Dataset Towards Point Cloud-based Organ-level Phenotyping of Sugar Beet Plants under Field Conditions [28.79416825695514]
  Agricultural production is facing severe challenges in the next decades induced by climate change and the need for sustainability.<n> Advancements in field management through non-chemical weeding by robots in combination with monitoring of crops by autonomous unmanned aerial vehicles (UAVs) are helpful to address these challenges.<n>The analysis of plant traits, called phenotyping, is an essential activity in plant breeding, it however involves a great amount of manual labor.
  arXiv  Detail & Related papers  (2023-12-22T14:06:44Z)
• Multi-growth stage plant recognition: a case study of Palmer amaranth (Amaranthus palmeri) in cotton (Gossypium hirsutum) [0.3441021278275805]
  We investigate eight-class growth stage recognition of Amaranthus palmeri in cotton. We compare 26 different architecture variants from YOLO v3, v5, v6, v6 3.0, v7, and v8. Highest mAP@[0.5:0.95] for recognition of all growth stage classes was 47.34% achieved by v8-X.
  arXiv  Detail & Related papers  (2023-07-28T21:14:43Z)
• Estimating Crop Primary Productivity with Sentinel-2 and Landsat 8 using Machine Learning Methods Trained with Radiative Transfer Simulations [58.17039841385472]
  We take advantage of all parallel developments in mechanistic modeling and satellite data availability for advanced monitoring of crop productivity. Our model successfully estimates gross primary productivity across a variety of C3 crop types and environmental conditions even though it does not use any local information from the corresponding sites. This highlights its potential to map crop productivity from new satellite sensors at a global scale with the help of current Earth observation cloud computing platforms.
  arXiv  Detail & Related papers  (2020-12-07T16:23:13Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.