YOLO11优化：多尺度提取能力 | 多尺度注意力网络（HSAN）通过分组卷积和联合通道-空间注意力机制，增强多尺度特征表达能力，2025年8月发布

 class HSAN(nn.Module):
    def __init__(self, c1, c2, shortcut=True, g=4, reduction=16):
        super().__init__()

        # 卷积层
        self.cv1 = nn.Conv2d(c1, c2, 1, 1, groups=g)
        self.cv2 = nn.Conv2d(c2, c2, 1, 1, groups=g)

        # 特殊张量分解
        self.num_groups = g
        self.group_channels = c2 // g

        # 全局池化
        self.avg_pool_h = nn.AdaptiveAvgPool2d((1, 1))
        self.max_pool_h = nn.AdaptiveMaxPool2d((1, 1))
        self.avg_pool_w = nn.AdaptiveAvgPool2d((1, 1))
        self.max_pool_w = nn.AdaptiveMaxPool2d((1, 1))

        # 多尺度卷积模块
        self.multi_scale_attention = nn.ModuleList([
            self.MSCM(self.group_channels) 
            for _ in range(g)
        ])

        # 共享卷积层
        self.channel_attention = nn.ModuleList([
            nn.Sequential(
                nn.Conv2d(self.group_channels, self.group_channels // reduction, 1, 1),
                nn.ReLU(inplace=True),
                nn.Conv2d(self.group_channels // reduction, self.group_channels, 1, 1)
            ) for _ in range(g)
        ])

        # 注意力权重计算
        self.coord_reduce_h = nn.Conv2d(2, 1, kernel_size=7, padding=3)
        self.coord_reduce_w = nn.Conv2d(2, 1, kernel_size=7, padding=3)

        self.sigmoid_h = nn.Sigmoid()
        self.sigmoid_w = nn.Sigmoid()
        self.sigmoid_channel = nn.Sigmoid()

        # 注意力应用
        self.shortcut = shortcut and c1 == c2

 # Ultralytics YOLO 🚀, AGPL-3.0 license
# YOLO11 object detection model with P3-P5 outputs. For Usage examples see https://docs.ultralytics.com/tasks/detect

# Parameters
nc: 80 # number of classes
scales: # model compound scaling constants, i.e. 'model=yolo11n.yaml' will call yolo11.yaml with scale 'n'
  # [depth, width, max_channels]
  n: [0.50, 0.25, 1024] # summary: 319 layers, 2624080 parameters, 2624064 gradients, 6.6 GFLOPs
  s: [0.50, 0.50, 1024] # summary: 319 layers, 9458752 parameters, 9458736 gradients, 21.7 GFLOPs
  m: [0.50, 1.00, 512] # summary: 409 layers, 20114688 parameters, 20114672 gradients, 68.5 GFLOPs
  l: [1.00, 1.00, 512] # summary: 631 layers, 25372160 parameters, 25372144 gradients, 87.6 GFLOPs
  x: [1.00, 1.50, 512] # summary: 631 layers, 56966176 parameters, 56966160 gradients, 196.0 GFLOPs

# YOLO11n backbone
backbone:
  # [from, repeats, module, args]
  - [-1, 1, Conv, [64, 3, 2]] # 0-P1/2
  - [-1, 1, Conv, [128, 3, 2]] # 1-P2/4
  - [-1, 2, C3k2, [256, False, 0.25]]
  - [-1, 1, Conv, [256, 3, 2]] # 3-P3/8
  - [-1, 2, C3k2, [512, False, 0.25]]
  - [-1, 1, Conv, [512, 3, 2]] # 5-P4/16
  - [-1, 2, C3k2, [512, True]]
  - [-1, 1, Conv, [1024, 3, 2]] # 7-P5/32
  - [-1, 2, C3k2, [1024, True]]
  - [-1, 1, SPPF, [1024, 5]] # 9
  - [-1, 2, HSAN, [1024]] # 10

# YOLO11n head
head:
  - [-1, 1, nn.Upsample, [None, 2, "nearest"]]
  - [[-1, 6], 1, Concat, [1]] # cat backbone P4
  - [-1, 2, C3k2, [512, False]] # 13

  - [-1, 1, nn.Upsample, [None, 2, "nearest"]]
  - [[-1, 4], 1, Concat, [1]] # cat backbone P3
  - [-1, 2, C3k2, [256, False]] # 16 (P3/8-small)

  - [-1, 1, Conv, [256, 3, 2]]
  - [[-1, 13], 1, Concat, [1]] # cat head P4
  - [-1, 2, C3k2, [512, False]] # 19 (P4/16-medium)

  - [-1, 1, Conv, [512, 3, 2]]
  - [[-1, 10], 1, Concat, [1]] # cat head P5
  - [-1, 2, C3k2, [1024, True]] # 22 (P5/32-large)

  - [[16, 19, 22], 1, Detect, [nc]] # Detect(P3, P4, P5)