Table 5 MobileNetV3 network parameters.
From: Detection of kidney bean leaf spot disease based on a hybrid deep learning model
Input size | Related operations | Dilation rate | Output channels | SE | NL | s |
|---|---|---|---|---|---|---|
2242 × 3 | Conv2d |  | 16 |  | HS | 2 |
1122 × 16 | Bneck,3 × 3 | 16 | 16 |  | RE | 1 |
1122 × 16 | Bneck,3 × 3 | 64 | 24 |  | RE | 2 |
562 × 24 | Bneck,3 × 3 | 72 | 24 |  | RE | 1 |
562 × 24 | Bneck,5 × 5 | 72 | 40 | \(\surd \) | RE | 2 |
282 × 40 | Bneck,5 × 5 | 120 | 40 | \(\surd \) | RE | 1 |
282 × 40 | Bneck,5 × 5 | 120 | 40 | \(\surd \) | RE | 1 |
282 × 40 | Bneck,3 × 3 | 240 | 80 |  | HS | 2 |
142 × 80 | Bneck,3 × 3 | 200 | 80 |  | HS | 1 |
142 × 80 | Bneck,3 × 3 | 184 | 80 |  | HS | 1 |
142 × 80 | Bneck,3 × 3 | 184 | 80 |  | HS | 1 |
142 × 80 | Bneck,3 × 3 | 480 | 112 | \(\surd \) | HS | 1 |
142 × 112 | Bneck,3 × 3 | 672 | 112 | \(\surd \) | HS | 1 |
142 × 112 | Bneck,5 × 5 | 672 | 160 | \(\surd \) | HS | 1 |
142 × 112 | Bneck,5 × 5 | 672 | 160 | \(\surd \) | HS | 2 |
72 × 160 | Bneck,5 × 5 | 960 | 160 | \(\surd \) | HS | 1 |
72 × 160 | Conv2d,1 × 1 |  | 960 |  | HS | 1 |
72 × 960 | Pool,7 × 7 |  |  |  | HS | 1 |
12 × 960 | Conv2d1 × 1, NBN |  | 1280 |  | HS | 1 |
12 × 1280 | Conv2d1 × 1, NBN |  | 1000 |  |  | 1 |
