Search claims · PlantClaims

Method / resourceobservedoriginal resultpositivesignificant (as stated)measured

Claim · model-written summary

YOLOV8s reached 121.5 FPS, faster than YOLOV5 at 102.7 FPS and Faster R-CNN at 11 FPS.

Taxon: TomatoSolanum lycopersicumt0
Outcome: frame ratee0
Direction: superiore0
Value: 121.5 FPSe0
Qualifier: Faster R-CNN’s 11 FPSe0
Comparator: YOLOV5e0

Verbatim

e0
“The YOLOV8s model has a significantly superior frame rate, reaching 121.5 FPS, in contrast to YOLOV5’s 102.7 FPS and Faster R-CNN’s 11 FPS”
✓ located verbatim in the source · abstract, characters 963–1101

t0
“Tomato Leaves”
✓ located verbatim in the source · title, characters 58–71

Title

A Deep-Learning-Based Model for the Detection of Diseased Tomato Leaves

Akram Abdullah; Gehad Abdullah Amran; S. M. Ahanaf Tahmid; et al. Agronomy 2024. doi:10.3390/agronomy14071593

Other claims from this paper (2)

Method / resource YOLOV8s attained 92.5% mean average precision for detecting diseased tomato leaves, exceeding YOLOV5 at 89.1% and Faster R-CNN at 77.5%.
Method / resource YOLOV8s was considerably smaller than the comparison models.

Method / resourceobservedoriginal resultmeasured

Claim · model-written summary

The pipeline reported 67.65 FPS for 67.65 FPS.

Taxon: cut chrysanthemumt0
Outcome: 67.65 FPSe0
Value: 67.65 FPSe0

Verbatim

e0
“fast inference time (14.78 ms per image), and 67.65 FPS”
✓ located verbatim in the source · abstract, characters 682–737

t0
“cut chrysanthemum”
✓ located verbatim in the source · title, characters 81–98

Title

Deep learning-based approach for extracting inflorescence morphology features in cut chrysanthemum

Shanpeng Xu; Jingshan Lu; Yin Wu; et al. Smart Agricultural Technology 2025. doi:10.1016/j.atech.2025.101265

Other claims from this paper (10)

Method / resource The pipeline reported 95.24% accuracy for flower type classification.
Method / resource The pipeline reported 1.26M parameters for parameters.
Method / resource The pipeline reported 0.15 GFLOPs for 0.15 GFLOPs.
Method / resource The pipeline reported 14.78 ms per image for inference time.
Method / resource The pipeline reported reduction of over 95% for parameters.
Method / resource The pipeline reported 0.934 for average Dice similarity coefficient (DSC).
Method / resource The pipeline reported 6.605 mm for mean squared errors (MSE).
Method / resource The pipeline reported 2.034 mm for mean squared errors (MSE).
Method / resource The pipeline reported approaching 0.98 for coefficients of determination (R2).
Method / resource The pipeline reported above 0.87 for F1-scores.

Method / resourceobservedoriginal resultpositivemeasured

Claim · model-written summary

The frame rate was 43.92 FPS versus 51.02 FPS for the original model.

Taxon: walnutt0
Outcome: frame ratee0
Value: 43.92 FPSe0
Qualifier: 51.02 FPSe0
Comparator: original modele0

Verbatim

e0
“The frame rate remains at 43.92 FPS, approaching the original model’s 51.02 FPS.”
✓ located verbatim in the source · abstract, characters 1347–1427

t0
“Walnut”
✓ located verbatim in the source · title, characters 0–6

Title

Walnut Surface Defect Classification and Detection Model Based on Enhanced YOLO11n

Xinyi Ma; Zhongjia Hao; Shuangyin Liu; et al. Agriculture 2025. doi:10.3390/agriculture15151707

Other claims from this paper (8)

Method / resource YOLO11-GME reached 96.2% mean average precision.
Method / resource YOLO11-GME improved mAP by 8.6% over YOLOv5n.
Method / resource YOLO11-GME improved mAP by 7% over YOLOv8n.
Method / resource YOLO11-GME improved mAP by 5.8% over YOLOv10n.
Method / resource YOLO11-GME improved mAP by 5.9% over YOLOv11.
Method / resource Normal walnut precision increased by 8.7%.
Method / resource Fissure precision increased by 5.3%.
Method / resource Inferior category precision increased by 3.7%.

Method / resourceobservedoriginal resultpositivemeasured

Claim · model-written summary

The detector ran at 154 FPS, outperforming rival detectors by 12–132 FPS.

Taxon: riceOryza sativat0
Outcome: frame ratee2
Value: 154 FPSe2
Qualifier: outperforming rivals by 12–132 FPSe2
Comparator: rivalse2

Verbatim

t0
“rice panicles”
✓ located verbatim in the source · title, characters 104–117

e2
“a frame rate of 154 FPS (outperforming rivals by 12–132 FPS)”
✓ located verbatim in the source · abstract, characters 1535–1595

Title

Dual attention guided context-aware feature learning for residual unfilled grains detection on threshed rice panicles

Yuhao Zhou; Xiao Feng; Shuqi Tang; et al. Artificial Intelligence in Agriculture 2025. doi:10.1016/j.aiia.2025.11.005

Other claims from this paper (1)

Method / resource The detector reached an mAP@0.5 of 95.3%, outperforming rival detectors by 1.5–32.6%.

Method / resourceobservedoriginal resultpositivemeasured

Claim · model-written summary

Inference speed rose from 30.36 FPS to 60.14 FPS.

Taxon: tea shootst0
Outcome: inference speede0
Direction: nearly doublese0
Value: 60.14 FPSe0
Qualifier: 30.36 FPSe0

Verbatim

e0
“nearly doubles inference speed from 30.36 FPS to 60.14 FPS”
✓ located verbatim in the source · abstract, characters 1390–1448

t0
“tea shoots”
✓ located verbatim in the source · title, characters 78–88

Title

Pluckable-zone segmentation and plucking-point localization for multicultivar tea shoots using an improved lightweight DeepLabV3+ and DBSCAN–assisted 3D clustering

Decheng Liu; Pengfei Wang; Zhi Zhang; et al. Information Processing in Agriculture 2026. doi:10.1016/j.inpa.2026.05.012

Other claims from this paper (11)

Method / resource The model used 6.04 M parameters.
Method / resource The model ran at 60.14 frames per second.
Method / resource Mean intersection over union was 81.77%.
Method / resource F1 score was 89.12%.
Method / resource Against original DeepLabV3+, mIoU improved by 10.78 percentage points.
Method / resource Parameter count fell from 54.71 M to 6.04 M.
Method / resource Cultivar mIoU ranged from 75.36% to 85.26%.
Method / resource Cultivar F1 scores ranged from 84.71% to 91.62%.
Method / resource Residual errors mainly came from illumination, occlusion and color contrast.
Method / resource Two dimensional localization had an 85.17% cultivar weighted success rate on 400 field images.
Method / resource DBSCAN based three dimensional processing achieved 92.00% plucking point accuracy in occluded and overlapping shoots.

Method / resourceobservedoriginal resultpositivemeasured

Claim · model-written summary

YOLOv8-VEW ran at 90 FPS, 24 FPS better than YOLOv8.

Taxon: SoybeanGlycine maxt0
Outcome: FPSe0
Direction: bettere0
Value: 90 FPSe0
Qualifier: 24 FPSe0
Comparator: YOLOv8 modele0

Verbatim

e0
“the F1, mAP, and FPS (frames per second) of the YOLOv8-VEW model were 0.95, 96.9%, and 90 FPS, respectively, which were 0.05, 2.4%, and 24 FPS better than those of the YOLOv8 model.”
✓ located verbatim in the source · abstract, characters 692–873

t0
“Soybean Flower and Pod Detection”
✓ located verbatim in the source · title, characters 12–44

Title

Field-Based Soybean Flower and Pod Detection Using an Improved YOLOv8-VEW Method

Kunpeng Zhao; Jinyang Li; Wenqiang Shi; et al. Agriculture 2024. doi:10.3390/agriculture14081423

Other claims from this paper (4)

Method / resource YOLOv8-VEW F1 was 0.95, 0.05 above YOLOv8.
Method / resource YOLOv8-VEW mAP was 96.9%, 2.4% better than YOLOv8.
Method / resource Model flower counts matched manual counts with R2 0.98311.
Method / resource Model pod counts matched manual counts with R2 0.98926.

Method / resourceobservedoriginal resultpositivesignificant (as stated)measured

Claim · model-written summary

On Nvidia Jetson Xavier NX, YOLOv8n-MCP achieved 49.5 FPS versus 37.6 FPS for YOLOv8n.

Taxon: strawberryFragaria ×ananassat0
Outcome: FPSe0
Direction: outperforminge0
Value: 49.5 FPSe0
Qualifier: 37.6 FPSe0
Comparator: original YOLOv8ne0
Setting: Nvidia Jetson Xavier NXe0

Verbatim

t0
“strawberry”
✓ located verbatim in the source · title, characters 36–46

e0
“Tests on Nvidia Jetson Xavier NX demonstrate that YOLOv8n-MCP achieves 49.5 FPS, significantly outperforming the original YOLOv8n’s 37.6 FPS”
✓ located verbatim in the source · abstract, characters 1125–1265

Title

Efficient instance segmentation for strawberry in greenhouses using YOLOv8n-MCP on edge devices

Xinhao Zhang; Guangpeng Zhang; Jiayi Wang; et al. Information Processing in Agriculture 2025. doi:10.1016/j.inpa.2025.07.001

Other claims from this paper (3)

Method / resource YOLOv8n-MCP reduced parameters by 69% relative to YOLOv8n.
Method / resource YOLOv8n-MCP reduced GFLOPs by 56% relative to YOLOv8n.
Method / resource YOLOv8n-MCP increased FPS by 42% relative to YOLOv8n.

Method / resourceobservedoriginal resultpositivemeasured

Claim · model-written summary

FCHE-YOLO inference speed rose to 47.5 FPS from 43.3 FPS.

Taxon: teat0
Outcome: inference speede0
Direction: increasese0
Value: 47.5 FPSe0
Comparator: 43.3 FPSe0

Verbatim

e0
“the inference speed of the model increases from 43.3 FPS to 47.5 FPS, with an increase of 9.0%”
✓ located verbatim in the source · abstract, characters 1349–1443