关键词: 空洞卷积, 上下文特征, 双重注意力机制, 密度图, 人群计数

Abstract: In response to challenges such as scale variations, background interference, and partial occlusion in crowd counting within complex scenes, a DA-DCCNN (Dual Attention based Dilated Contextual Convolutional Neural Network) was proposed. Firstly, the convolutional layers from VGG16 were utilized as feature extractors to obtain abstract and deep-level feature maps of crowd images. Subsequently, by employing dilated convolutions, a Dilation Context Module (DCM) was constructed to connect features obtained from different layers. The Spatial Attention Module (SAM) and Channel Attention Module (CAM) were introduced to acquire contextual information. Finally, a loss function was formulated by combining the Euclidean distance and cross-entropy to measure the disparity between the predicted attention map and the ground truth attention map. Experimental results on three publicly available datasets—ShanghaiTech, UCF_CC_50, and UCF-QNRF demonstrate that DA-DCCNN can effectively capture multi-scale features in images and enhance the perception of important regions and channels within the images. The Mean Absolute Error (MAE) achieves relatively optimal results. It can be seen that the feature fusion network based on the dual attention mechanism can efficiently recognize spatial structures and local features in images, leading to more accurate predictions and crowd area counts in the generated density maps.

Key words: dilated convolution, contextual feature, dual-attention mechanism, density map, crowd counting