Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Heterogeneous parallel computing system and method for threshold method image segmentation, and electronic equipment

An image segmentation and parallel computing technology, applied in the field of image processing, can solve the problems of low processing efficiency and lack of high-performance parallel computing system design, and achieve the effect of improving processing efficiency, maintaining control flexibility, and reducing data volume.

Pending Publication Date: 2022-02-11
HARBIN INST OF TECH
View PDF0 Cites 2 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] To sum up, the threshold method image segmentation is widely used in pre-processing such as target area pre-selection due to its simple calculation process, but its processing method of traversing the image pixel by pixel is less efficient under the serial processing capability provided by the edge microprocessor. Low, there is currently a lack of high-performance parallel computing system design for such methods

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Heterogeneous parallel computing system and method for threshold method image segmentation, and electronic equipment
  • Heterogeneous parallel computing system and method for threshold method image segmentation, and electronic equipment
  • Heterogeneous parallel computing system and method for threshold method image segmentation, and electronic equipment

Examples

Experimental program
Comparison scheme
Effect test

Embodiment

[0113] The present invention uses high-resolution remote sensing images as input to perform sea-land segmentation processing, and the size of the input image is 8192×2048 pixels. After entering the image slice function unit in the microprocessor, the original image will be sequentially divided into sub-images with a size of 512×512 pixels in the direction of rows first and then columns, and the sub-images will be transmitted to the FPGA through the interconnection bus between heterogeneous processors , for subsequent image segmentation processing.

[0114] The sliced ​​subgraph first enters the subgraph classification and coarse segmentation computing unit based on local features in the FPGA for processing. The present invention designs a subgraph classification process based on parallel local feature extraction to accelerate the calculation of the subgraph classification process, such as figure 2 shown. Divide the submap into 64 local windows, each containing 64×64 pixels....

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention provides a heterogeneous parallel computing system and method for threshold value method image segmentation, and electronic equipment. The system comprises an image slicing function unit, a sub-image classification coarse segmentation calculation unit based on local features, a target and background boundary sub-image fine segmentation calculation unit based on an OTSU threshold value and a sub-image segmentation result splicing function unit. On the basis of processing characteristic analysis of functional units of threshold method image segmentation, a microprocessor and an FPGA (Field Programmable Gate Array) provided by a heterogeneous system-on-chip are utilized to realize design of a heterogeneous computing system for controlling intensive tasks and computing intensive tasks; a heterogeneous system-on-chip has the characteristics of flexible control capability of a microprocessor and high computing capability of an FPGA (Field Programmable Gate Array), a threshold method image segmentation heterogeneous parallel computing system is designed, and control flexibility of rearrangement of image pixel positions is kept while accelerated computing is performed on processing steps which consume long computing time in a threshold method, and overall execution efficiency of the image segmentation method is improved.

Description

technical field [0001] The invention belongs to the technical field of image processing, and in particular relates to a heterogeneous parallel computing system, method and electronic equipment for image segmentation based on a threshold method. Background technique [0002] Image segmentation is one of the basic image processing links in computer vision tasks. It aims to divide a complete image into multiple sub-regions. There are large differences between different sub-regions, while the pixels in the same sub-region have similar characteristics, so as to achieve the goal. Separation of local area from background area. In image processing tasks with strong timeliness and high precision requirements such as target detection and target tracking, image segmentation is often used as an important preprocessing step to preselect the area where the target is located, reduce the amount of data, and improve the efficiency and accuracy of target detection. At present, methods such a...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
Patent Type & Authority Applications(China)
IPC IPC(8): G06T7/136G06T7/194G06K9/62
CPCG06T7/136G06T7/194G06T2207/20021G06F18/2415
Inventor 刘连胜彭宇于希明姚博文彭喜元
Owner HARBIN INST OF TECH
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products