Abstract
The realization that mobile phones can detect rice diseases and insect pests not only solves the problems of low efficiency and poor accuracy from manually detection and reporting, but it also helps farmers detect and control them in the field in a timely fashion, thereby ensuring the quality of rice grains. This study examined two Improved detection models for the detection of six high-frequency diseases and insect pests. These models were the Improved You Only Look Once (YOLO)v5s and YOLOv7-tiny based on their lightweight object detection networks. The Improved YOLOv5s was introduced with the Ghost module to reduce computation and optimize the model structure, and the Improved YOLOv7-tiny was introduced with the Convolutional Block Attention Module (CBAM) and SIoU to improve model learning ability and accuracy. First, we evaluated and analyzed the detection accuracy and operational efficiency of the models. Then we deployed two proposed methods to a mobile phone. We also designed an application to further verify their practicality for detecting rice diseases and insect pests. The results showed that Improved YOLOv5s achieved the highest F1-Score of 0.931, 0.961 in mean average precision (mAP) (0.5), and 0.648 in mAP (0.5:0.9). It also reduced network parameters, model size, and the floating point operations per second (FLOPs) by 47.5, 45.7, and 48.7%, respectively. Furthermore, it increased the model inference speed by 38.6% compared with the original YOLOv5s model. Improved YOLOv7-tiny outperformed the original YOLOv7-tiny in detection accuracy, which was second only to Improved YOLOv5s. The probability heat maps of the detection results showed that Improved YOLOv5s performed better in detecting large target areas of rice diseases and insect pests, while Improved YOLOv7-tiny was more accurate in small target areas. On the mobile phone platform, the precision and recall of Improved YOLOv5s under FP16 accuracy were 0.925 and 0.939, and the inference speed was 374 ms/frame, which was superior to Improved YOLOv7-tiny. Both of the proposed improved models realized accurate identification of rice diseases and insect pests. Moreover, the constructed mobile phone application based on the improved detection models provided a reference for realizing fast and efficient field diagnoses.
Abstract from the paper’s OpenAlex record; open copy licensed cc-by. 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 15
Claim · model-written summary
Improved YOLOv5s reached 0.931 F1-Score.
- Taxon
- Rice
Oryza sativat0 - Outcome
- F1-Scoree0
- Direction
- higheste0
- Value
- 0.931e0
Organisms pictured · Representative pictures of the matched species, not figures from the paper.
Verbatim
t0“Rice”
e0“Improved YOLOv5s achieved the highest F1-Score of 0.931, 0.961 in mean average precision (mAP) (0.5), and 0.648 in mAP (0.5:0.9).”
Claim · model-written summary
Improved YOLOv5s reached 0.961 mean average precision (mAP) (0.5).
- Taxon
- RiceOryza sativat0
- Outcome
- mean average precision (mAP) (0.5)e0
- Direction
- higheste0
- Value
- 0.961e0
Organisms pictured · Representative pictures of the matched species, not figures from the paper.
Verbatim
t0“Rice”
e0“Improved YOLOv5s achieved the highest F1-Score of 0.931, 0.961 in mean average precision (mAP) (0.5), and 0.648 in mAP (0.5:0.9).”
Claim · model-written summary
Improved YOLOv5s reached 0.648 mAP (0.5:0.9).
- Taxon
- RiceOryza sativat0
- Outcome
- mAP (0.5:0.9)e0
- Direction
- higheste0
- Value
- 0.648e0
Organisms pictured · Representative pictures of the matched species, not figures from the paper.
Verbatim
t0“Rice”
e0“Improved YOLOv5s achieved the highest F1-Score of 0.931, 0.961 in mean average precision (mAP) (0.5), and 0.648 in mAP (0.5:0.9).”
Claim · model-written summary
Improved YOLOv5s reduced network parameters by 47.5.
- Taxon
- RiceOryza sativat0
- Outcome
- network parameterse0
- Direction
- reducede0
- Value
- 47.5e0
Organisms pictured · Representative pictures of the matched species, not figures from the paper.
Verbatim
t0“Rice”
e0“It also reduced network parameters, model size, and the floating point operations per second (FLOPs) by 47.5, 45.7, and 48.7%, respectively.”
Claim · model-written summary
Improved YOLOv5s reduced model size by 45.7.
- Taxon
- RiceOryza sativat0
- Outcome
- model sizee0
- Direction
- reducede0
- Value
- 45.7e0
Organisms pictured · Representative pictures of the matched species, not figures from the paper.
Verbatim
t0“Rice”
e0“It also reduced network parameters, model size, and the floating point operations per second (FLOPs) by 47.5, 45.7, and 48.7%, respectively.”
Claim · model-written summary
Improved YOLOv5s reduced floating point operations per second (FLOPs) by 48.7%.
- Taxon
- RiceOryza sativat0
- Outcome
- floating point operations per second (FLOPs)e0
- Direction
- reducede0
- Value
- 48.7%e0
Organisms pictured · Representative pictures of the matched species, not figures from the paper.
Verbatim
t0“Rice”
e0“It also reduced network parameters, model size, and the floating point operations per second (FLOPs) by 47.5, 45.7, and 48.7%, respectively.”
Claim · model-written summary
Improved YOLOv5s increased inference speed by 38.6% versus original YOLOv5s.
- Taxon
- RiceOryza sativat0
- Outcome
- model inference speede0
- Direction
- increasede0
- Value
- 38.6%e0
- Comparator
- original YOLOv5s modele0
Organisms pictured · Representative pictures of the matched species, not figures from the paper.
Verbatim
t0“Rice”
e0“it increased the model inference speed by 38.6% compared with the original YOLOv5s model.”
Claim · model-written summary
Improved YOLOv7-tiny outperformed its original version in detection accuracy but ranked below Improved YOLOv5s.
- Taxon
- RiceOryza sativat0
- Outcome
- detection accuracye0
- Direction
- outperformede0
- Comparator
- original YOLOv7-tinye0
Organisms pictured · Representative pictures of the matched species, not figures from the paper.
Verbatim
t0“Rice”
e0“Improved YOLOv7-tiny outperformed the original YOLOv7-tiny in detection accuracy, which was second only to Improved YOLOv5s.”
Claim · model-written summary
Improved YOLOv5s performed better on large rice disease and pest targets.
- Taxon
- riceOryza sativae0
- Outcome
- detecting large target areas of rice diseases and insect pestse0
- Direction
- bettere0
- Comparator
- Improved YOLOv7-tinye0
Organisms pictured · Representative pictures of the matched species, not figures from the paper.
Verbatim
e0“Improved YOLOv5s performed better in detecting large target areas of rice diseases and insect pests, while Improved YOLOv7-tiny was more accurate in small target areas.”
Claim · model-written summary
Improved YOLOv7-tiny was more accurate on small targets.
- Taxon
- riceOryza sativae0
- Outcome
- accurate in small target arease0
- Direction
- moree0
- Comparator
- Improved YOLOv5se0
Organisms pictured · Representative pictures of the matched species, not figures from the paper.
Verbatim
e0“Improved YOLOv5s performed better in detecting large target areas of rice diseases and insect pests, while Improved YOLOv7-tiny was more accurate in small target areas.”
Claim · model-written summary
On mobile FP16, Improved YOLOv5s achieved 0.925 precision.
- Taxon
- RiceOryza sativat0
- Outcome
- precisione0
- Value
- 0.925e0
- Comparator
- Improved YOLOv7-tinye0
- Setting
- mobile phone platforme0
Organisms pictured · Representative pictures of the matched species, not figures from the paper.
Verbatim
t0“Rice”
e0“On the mobile phone platform, the precision and recall of Improved YOLOv5s under FP16 accuracy were 0.925 and 0.939, and the inference speed was 374 ms/frame, which was superior to Improved YOLOv7-tiny.”
Claim · model-written summary
On mobile FP16, Improved YOLOv5s achieved 0.939 recall.
- Taxon
- RiceOryza sativat0
- Outcome
- recalle0
- Value
- 0.939e0
- Comparator
- Improved YOLOv7-tinye0
- Setting
- mobile phone platforme0
Organisms pictured · Representative pictures of the matched species, not figures from the paper.
Verbatim
t0“Rice”
e0“On the mobile phone platform, the precision and recall of Improved YOLOv5s under FP16 accuracy were 0.925 and 0.939, and the inference speed was 374 ms/frame, which was superior to Improved YOLOv7-tiny.”
Claim · model-written summary
On mobile FP16, Improved YOLOv5s achieved 374 ms/frame inference speed.
- Taxon
- RiceOryza sativat0
- Outcome
- inference speede0
- Value
- 374 ms/framee0
- Comparator
- Improved YOLOv7-tinye0
- Setting
- mobile phone platforme0
Organisms pictured · Representative pictures of the matched species, not figures from the paper.
Verbatim
t0“Rice”
e0“On the mobile phone platform, the precision and recall of Improved YOLOv5s under FP16 accuracy were 0.925 and 0.939, and the inference speed was 374 ms/frame, which was superior to Improved YOLOv7-tiny.”
Claim · model-written summary
Both improved models accurately identified rice diseases and insect pests.
- Taxon
- riceOryza sativae0
- Outcome
- accurate identification of rice diseases and insect pestse0
- Value
- accuratee0
Organisms pictured · Representative pictures of the matched species, not figures from the paper.
Verbatim
e0“Both of the proposed improved models realized accurate identification of rice diseases and insect pests.”
Claim · model-written summary
The mobile application provided a reference for fast field diagnosis.
- Taxon
- RiceOryza sativat0
- Outcome
- fast and efficient field diagnosese0
- Value
- faste0
Organisms pictured · Representative pictures of the matched species, not figures from the paper.
Verbatim
t0“Rice”
e0“the constructed mobile phone application based on the improved detection models provided a reference for realizing fast and efficient field diagnoses.”