A Deconvolutional Neural Network Training Method
A neural network training and deconvolution technology, applied in the field of deconvolution neural network training, to achieve the effects of improving training convergence efficiency and convergence accuracy, improving classification accuracy, and reducing training costs
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment Construction
[0023] Such as Figure 4 As shown, this deconvolutional neural network training method includes a training phase and a reconstruction phase, and the training phase includes the following steps:
[0024] (1) Preprocess the training image: select the training set image, process it as a grayscale image, and unify the length and width pixels;
[0025] (2) Batch setting is performed on the training images: according to the application of the trained network, the training images are divided into batches;
[0026] (3) Set the network training parameters of the training image. The network training parameters of the training image include the number of network layers, the filter size of each layer, the number of feature maps of each layer, the number of FISTA reconstruction steps and reconstruction steps, the total number of epoch cycles, Feature map sparse control parameters;
[0027] (4) Start the first layer training: initialize the first layer feature map and the first layer filt...
PUM
Login to View More Abstract
Description
Claims
Application Information
Login to View More 


