Abstract Recently, the application of attention mechanisms in convolutional neural networks (CNNs) has become a hot area in computer vision. Most existing methods focus on channel attention or spatial attention. Some mixed attention usually achieves better performance than channel attention or spatial attention with the help of a complex model structure, which increases the complexity of the model. This article proposes an efficient mixed attention that combines channel information with spatial information using learnable broadcast addition to reduce this complexity. In particular, this module can simplify learning and improve performance with fewer parameters. Furthermore, our method uses an excitation method based on the Tanh function to reduce computational resources while maintaining model performance, and it is a lightweight attention module that can be used in arbitrary CNNs to improve performance. Experiments on ImageNet and Cifar confirm the effectiveness of the proposed method. Besides, our method remains highly competitive for object detection tasks and image segmentation tasks.

In this paper, a resonant dc-dc converter with modular rectifier is proposed for high voltage gain and wide output voltage range applications such as interconnecting renewables to the medium voltage dc grid. The modular rectifier on the secondary side is used to support the high voltage and the number of submodules (SMs) can be chosen depending on the output voltage. Wide range of voltage regulation is achieved by regulating the frequency and the duty cycle of the SM gate signals. The proposed converter can achieve ZVS for the primary side switches and ZCS for the secondary rectifier. An 800W prototype with one SM in each arm is constructed to demonstrate the operating principle of the proposed converter. The efficiency under voltage gain 1, 1.25 and 1.5 are measured and the peak efficiency is 96.8%.