LDCNet: a lightweight multi-scale convolutional neural network using local dense connectivity for image recognition

Deep convolutional neural networks (DCNNs) have made great contributions to the development of computer vision. Since the trained DCNN models require a large amount of computing and storage resources to achieve high performance, it is usually difficult to deploy them on resource-limited systems. To address this problem, we propose a novel local dense connectivity (LDC) module to generate feature maps from cheap convolution operations. The LDC module constructs hierarchical and locally dense connections within a single layer. This construction promotes the reuse of features in network layers. As a result, it leads to the generation of multi-scale feature maps and increases the receptive fields for each network layer. A basic architecture block called LDCBlock was designed based on the LDC module. By stacking this kind of block, we propose a kind of lightweight and efficient multi-scale residual network named LDCNet. Moreover, to model the interdependencies between the channels of the LDC module and enhance the informativeness of diversified features, we also design a sparse Squeeze-Excitation (SE) module, which has fewer parameters and computations. Finally, the experiments based on CIFAR datasets, ImageNet dataset and a defect detection dataset demonstrate that our LDCNet achieves competitive performance compared with the state-of-the-art models focusing on model compression.