DA-ActNN-YOLOV5: Hybrid YOLO v5 Model with Data Augmentation and Activation of Compression Mechanism for Potato Disease Identification

Table 5

The methodology of this study compared to other research methods.

No.	References	Year	Method description	AP (%)
1	Sholihati et al. [32]	2020	VGGNet16 + VGGNet19	91.31
2	Yang et al. [19]	2020	Faster R-CNN + SIFT + K-means	90.83
3	Barman et al. [33]	2021	Self-build CNN (SBCNN)	96.98
4	D. F. Wang, and J. Wang [18]	2021	SE + ResNet50+DenseNet-121	97.99
5	Rashid et al. [34]	2021	PDDCNN + data augmentation	99.75
6	Afzaal et al. [35]	2021	GoogleNet + VGGNet + EfficientNet	94.00
7	Hou et al. [36]	2022	k-NN + SVM + ANN + RF	97.40
8	Chen et al. [37]	2022	MobileNetV2+GAN + Attention mechanism + octave	97.73
9	Mahum et al. [38]	2022	DenseNet-201 + Efficient DenseNet	97.20
10	Sharma et al. [39]	2022	Deep CNN	97.66
11	Proposed model	2022	YOLO v5+ data augmentation + ActNN (DA-ActNN-YOLOV5)	99.81