Moving target detection method based on deep optical flow and morphological method
A technology for moving targets and detection methods, applied in image data processing, image enhancement, instruments, etc., can solve the problems of incomplete detection target shape, low detection accuracy, etc., achieve strong learning ability, good detection robustness, and calculation effect Good results
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment Construction
[0068] The present invention will be described in detail below in conjunction with the accompanying drawings and embodiments.
[0069] In the present invention, the realization and verification of the moving target detection model concept is to use GPU (GTX1080) as the computing platform, adopt the GPU parallel computing framework, and select Caffe as the CNN (convolutional network) framework.
[0070] Such as figure 1 Shown, the steps of the present invention are: (1) collect video data, label sample video, and be divided into training set and test set at random, to the training set that has been processed and test set, do mean value calculation, form training set mean value file and test Set the mean value file to complete the preprocessing of the training set and test set; (2) Construct a fully convolutional neural network architecture, which consists of two parts, encoding and decoding, and use the training set and test set to train through an adaptive learning rate adjust...
PUM
Abstract
Description
Claims
Application Information
- R&D Engineer
- R&D Manager
- IP Professional
- Industry Leading Data Capabilities
- Powerful AI technology
- Patent DNA Extraction
Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.
© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com