ABSTRACT
Automatic and fast segmentation of retinal vessels in fundus images is a prerequisite in clinical ophthalmic diseases; however, the high model complexity and low segmentation accuracy still limit its application. This paper proposes a lightweight dual-path cascaded network (LDPC-Net) for automatic and fast vessel segmentation. We designed a dual-path cascaded network via two U-shaped structures. Firstly, we employed a structured discarding (SD) convolution module to alleviate the over-fitting problem in both codec parts. Secondly, we introduced the depthwise separable convolution (DSC) technique to reduce the parameter amount of the model. Thirdly, a residual atrous spatial pyramid pooling (ResASPP) model is constructed in the connection layer to aggregate multi-scale information effectively. Finally, we performed comparative experiments on three public datasets. Experimental results show that the proposed method achieved superior performance on the accuracy, connectivity, and parameter quantity, thus proving that it can be a promising lightweight assisted tool for ophthalmic diseases.
ABSTRACT
We present a two-dimensional (2D) parity-time-symmetric (PT-symmetry) phononic crystals (PCs) with balanced gain and loss medium. Using the super cell method of rectangular lattice, we exhibit the thresholdless spontaneous PT-symmetry breaking in the band structure. The numerical results show that the asymmetric scattering properties obviously occur in a non-Hermitian system. At two specific incident frequencies, unidirectional reflectionless and perfect transmission behaviors exist individually in opposite directions, which are accompanied by a phase transition of π. Based on the generalized Snell's law, combining such a PT-symmetric medium, we design a novel metamaterial crystal for PT-symmetric acoustic flat focusing. Its focus frequency can also be modulated by the gain/loss parameter. The novel flat focusing based on the PT-symmetry that we propose opens a new door for high-dimensional applications of non-Hermitian metamaterials in acoustic wave manipulation.