A dilation and erosion processing method for grayscale and binary images based on fpga

A binary image and erosion processing technology, applied in the field of image processing, can solve problems such as limited performance, memory bandwidth, etc., and achieve the effect of reducing the number of accesses and accelerating calculations

Active Publication Date: 2022-03-22
HEFEI CHIP FOUND MICROELECTRONICS EQUIP CO LTD
View PDF3 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

In the existing technology, expansion and erosion need to read the target element covered by the structural element multiple times, and the performance is limited by the memory bandwidth.

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
  • A dilation and erosion processing method for grayscale and binary images based on fpga
  • A dilation and erosion processing method for grayscale and binary images based on fpga
  • A dilation and erosion processing method for grayscale and binary images based on fpga

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0033] Select the grayscale image with a structural element window size of 3×3, a grayscale of 2, and a width of 32768 as the first embodiment, such as figure 1 , figure 2 , image 3 , Figure 4 , Figure 5 , Figure 6 and Figure 7 As shown, the process steps are described as follows:

[0034] Step 1: Set dilation or erosion, and determine the calculation method, which is used to specify the intra-row calculation and inter-row calculation methods.

[0035] Step 2: Copy the boundary pixels according to the window size of the structural element. The size of the structural window is 3×3, so the number of copied boundary pixels is 1; the boundary copy is realized by buffering the 2-shot register cache. The schematic diagram of the boundary copy is as follows Figure 4 shown. When copying border pixels, the first and last border pixels copy their own border pixels, and the middle border pixels copy adjacent border pixels.

[0036] Step 3: Perform in-line calculation on th...

Embodiment 2

[0040] Select the binary image with a structural element window size of 3×3 and a width of 32768 as the second embodiment, such as figure 1 , figure 2 , image 3 , Figure 8 , Figure 9 and Figure 10 As shown, the process steps are described as follows:

[0041] Step 1: Set dilation or erosion, and determine the calculation method, which is used to specify the intra-row calculation and inter-row calculation methods.

[0042] Step 2: Copy the boundary pixels according to the structuring element window size. Embodiment 2 The size of the structural window is 3×3, so the number of copied boundary pixels is 1. Embodiment 2 Realize boundary copy by caching 2 beat register cache, the schematic diagram of boundary copy is as follows Figure 8 shown. When copying border pixels, the first and last border pixels copy their own border pixels, and the middle border pixels copy adjacent border pixels.

[0043] Step 3: Perform in-row calculation on the data after copying the bound...

Embodiment 1

[0046] Embodiment 1 and Embodiment 2 have the same implementation block diagram. The implementation block diagram adopts a pipeline structure, and the data enters the boundary copy, intra-row calculation, row cache and inter-row calculation unit components in turn, and the selector controls the intra-row calculation and inter-row calculation units respectively. The size of the structural element window is optional. The present invention takes the window size of 3*3 as an example for illustration, which can be reflected in figure 2 The number of middle row caches is 3, Figure 6 The number of instantiated RAM is 3, Figure 4 and Figure 8 The number of border pixels assigned in the middle is 1, that is (3-1) / 2, Figure 5 , Figure 7 , Figure 9 and Figure 10 3 pixels are used as the calculation unit, but the structural element window is not drawn separately.

[0047] In summary, the present invention duplicates the left and right borders, and the width of the image aft...

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 relates to an FPGA-based dilation and erosion processing method for grayscale and binary images, which comprises the steps in the following sequence: setting expansion or erosion, determining the calculation method; copying the left and right boundary pixels according to the window width of the structural element; The erosion algorithm performs in-line calculation on the data after copying the left and right boundary pixels; uses the FPGA RAM to implement a shift register to cache the calculation results in the line; when the calculation results in the cache line reach the height of a structural element window, read the data in the cache RAM at the same time, And again according to the dilation and corrosion algorithm, calculate and output the cached data between rows. The present invention duplicates the left and right borders, and the width of the expanded and eroded image is consistent with the original image width; the present invention effectively reduces the number of visits to the target elements covered by the structural element window through line buffering and pipeline, and realizes the accelerated calculation of dilation and erosion.

Description

technical field [0001] The invention relates to the technical field of image processing, in particular to an FPGA-based dilation and erosion processing method for grayscale and binary images. Background technique [0002] Based on set theory, mathematical morphology defines a set of transformations based on the set of structural elements. It calculates the transformation of target elements and structural elements to obtain the shape of the object, so as to achieve the purpose of image analysis and recognition. [0003] Dilation and erosion are fundamental transformations in mathematical morphology, defined as the union and intersection of a structure element and a target element, respectively. For binary images, that is, images with pixel values ​​of 0 and 1, expansion and erosion correspond to the OR operation and AND operation of the structure element covering the target element; for grayscale images, that is, the pixel values ​​​​are finite non-negative integers, expansio...

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 Patents(China)
IPC IPC(8): G06T7/155G06T1/60
CPCG06T1/60G06T7/155
Inventor 卞洪飞
Owner HEFEI CHIP FOUND MICROELECTRONICS EQUIP CO LTD
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap