Construction Method of Kaleidoscope Convolutional Network

Abstract

A method for constructing a kaleidoscope convolutional network belongs to a deep learning algorithm network. The present invention aims at the problem that the existing hole filter is used to expand the receptive field of the convolutional network, and the boundary effect generated when the hole rate is set too large affects the feature capture. Including: constructing a kaleidoscope filter, making the outer contour of the kaleidoscope filter a square, dividing the square into four division units, and determining nine first-level sampling points by the four corners of the four division units; In the middle, three secondary sampling points are determined by the intersection positions of kaleidoscope petals; then a kaleidoscope cascade spatial pyramid pooling module is constructed based on a plurality of kaleidoscope filters of different sizes; finally, Xception is used as the backbone network, and the kaleidoscope level Linked spatial pyramid pooling modules are assembled at the decoder layer to form a kaleidoscope convolutional network. The present invention can better capture the shape features of the object.

Description

technical field [0001] The invention relates to a construction method of a kaleidoscope convolutional network, which belongs to a deep learning algorithm network. Background technique [0002] Deep feature representations based on convolutional neural networks (CNNs) have greatly improved semantic segmentation. At present, more successful semantic segmentation methods based on CNNs mainly rely on high-resolution representations and rich feature representations. [0003] It is generally believed that the larger the convolution size, the larger the receptive field, and the more accurate the features captured. These features include texture features, shape features, and contextual features, which are crucial for object recognition. But larger convolution filters lead to higher computational cost. [0004] In order to obtain a large receptive field with low computational cost, many traditional convolutional networks use a large number of 3×3 filters for serial convolution, an...

AI Technical Summary

Problems solved by technology

[0006] Aiming at the problem that the existing hole filter is used to expand the receptive field of the convolutional network, and the boundary effect generated when the hole rate is set too large affects the feature capture, the present invention provides a method for constructing a kaleidoscope convolutional network

Images