Abstract
BACKGROUND: Ripeness is a phenotype that significantly impacts the quality of fruits, constituting a crucial factor in the cultivation and harvesting processes. Manual detection methods and experimental analysis, however, are inefficient and costly. RESULTS: In this study, we propose a lightweight and efficient melon ripeness detection method, MRD-YOLO, based on an improved object detection algorithm. The method combines a lightweight backbone network, MobileNetV3, a design paradigm Slim-neck, and a Coordinate Attention mechanism. Additionally, we have created a large-scale melon dataset sourced from a greenhouse based on ripeness. This dataset contains common complexities encountered in the field environment, such as occlusions, overlapping, and varying light intensities. MRD-YOLO achieves a mean Average Precision of 97.4% on this dataset, achieving accurate and reliable melon ripeness detection. Moreover, the method demands only 4.8 G FLOPs and 2.06 M parameters, representing 58.5% and 68.4% of the baseline YOLOv8n model, respectively. It comprehensively outperforms existing methods in terms of balanced accuracy and computational efficiency. Furthermore, it maintains real-time inference capability in GPU environments and demonstrates exceptional inference speed in CPU environments. The lightweight design of MRD-YOLO is anticipated to be deployed in various resource constrained mobile and edge devices, such as picking robots. Particularly noteworthy is its performance when tested on two melon datasets obtained from the Roboflow platform, achieving a mean Average Precision of 85.9%. This underscores its excellent generalization ability on untrained data. CONCLUSIONS: This study presents an efficient method for melon ripeness detection, and the dataset utilized in this study, alongside the detection method, will provide a valuable reference for ripeness detection across various types of fruits.
Abstract from the paper’s OpenAlex record; open copy licensed cc-by-nc-nd. Read at the source ↗
The text the model read, with the source record’s inline markup (italics, sub- and superscripts) rendered and its character entities decoded; quote offsets refer to the raw text.
Claims 7
Claim · model-written summary
A greenhouse-sourced melon ripeness dataset was created with occlusion, overlap and lighting variation.
- Taxon
- melon
.jpg?width=60)
Cucumis meloe0 - Outcome
- melon datasete0
- Setting
- greenhousee0
Not reportedComparatorValue
Organisms pictured · Representative pictures of the matched species, not figures from the paper.
.jpg?width=500)
Verbatim
e0“we have created a large-scale melon dataset sourced from a greenhouse based on ripeness. This dataset contains common complexities encountered in the field environment, such as occlusions, overlapping, and varying light intensities.”
Claim · model-written summary
MRD-YOLO achieved 97.4% mean Average Precision on the melon dataset.
- Taxon
- melonCucumis meloe0
- Outcome
- mean Average Precisione0
- Direction
- achievinge0
- Value
- 97.4%e0
Organisms pictured · Representative pictures of the matched species, not figures from the paper.
Verbatim
e0“MRD-YOLO achieves a mean Average Precision of 97.4% on this dataset, achieving accurate and reliable melon ripeness detection.”
Claim · model-written summary
MRD-YOLO used 4.8 G FLOPs, 58.5% of the YOLOv8n baseline.
Verbatim
e0“the method demands only 4.8 G FLOPs and 2.06 M parameters, representing 58.5% and 68.4% of the baseline YOLOv8n model, respectively.”
Claim · model-written summary
MRD-YOLO used 2.06 M parameters, 68.4% of the YOLOv8n baseline.
Verbatim
e0“the method demands only 4.8 G FLOPs and 2.06 M parameters, representing 58.5% and 68.4% of the baseline YOLOv8n model, respectively.”
Claim · model-written summary
MRD-YOLO outperformed existing methods in balanced accuracy and computational efficiency.
Verbatim
e0“It comprehensively outperforms existing methods in terms of balanced accuracy and computational efficiency.”
Claim · model-written summary
MRD-YOLO retained real-time GPU inference and exceptional CPU inference speed.
Not reportedComparatorValue
Verbatim
e0“it maintains real-time inference capability in GPU environments and demonstrates exceptional inference speed in CPU environments.”
Claim · model-written summary
MRD-YOLO reached 85.9% mean Average Precision on two unseen Roboflow melon datasets.
- Taxon
- melonCucumis meloe0
- Outcome
- mean Average Precisione0
- Value
- 85.9%e0
- Qualifier
- untrained datae0
Organisms pictured · Representative pictures of the matched species, not figures from the paper.
Verbatim
e0“when tested on two melon datasets obtained from the Roboflow platform, achieving a mean Average Precision of 85.9%. This underscores its excellent generalization ability on untrained data.”