CN106570230A - Robinson filtering design method based on FPGA (Field Programmable Gate Array) - Google Patents
Robinson filtering design method based on FPGA (Field Programmable Gate Array) Download PDFInfo
- Publication number
- CN106570230A CN106570230A CN201610919283.9A CN201610919283A CN106570230A CN 106570230 A CN106570230 A CN 106570230A CN 201610919283 A CN201610919283 A CN 201610919283A CN 106570230 A CN106570230 A CN 106570230A
- Authority
- CN
- China
- Prior art keywords
- row
- robinson
- fifo
- clock
- data
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/30—Circuit design
-
- H—ELECTRICITY
- H03—ELECTRONIC CIRCUITRY
- H03H—IMPEDANCE NETWORKS, e.g. RESONANT CIRCUITS; RESONATORS
- H03H17/00—Networks using digital techniques
- H03H17/02—Frequency selective networks
- H03H17/0248—Filters characterised by a particular frequency response or filtering method
- H03H17/0261—Non linear filters
-
- H—ELECTRICITY
- H03—ELECTRONIC CIRCUITRY
- H03H—IMPEDANCE NETWORKS, e.g. RESONANT CIRCUITS; RESONATORS
- H03H17/00—Networks using digital techniques
- H03H2017/0072—Theoretical filter design
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Computer Hardware Design (AREA)
- Theoretical Computer Science (AREA)
- Nonlinear Science (AREA)
- Mathematical Physics (AREA)
- Evolutionary Computation (AREA)
- Geometry (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Image Processing (AREA)
Abstract
The invention relates to a Robinson filtering design method based on an FPGA (Field Programmable Gate Array). Six FIFO (First In First Out) memories are sequentially connected, input data indatabus is read out sequentially through the six FIFOs, each piece of clock input data and data read out by six FIFOs form seven rows of data in the same line and with adjacent images, seven pieces of data sequentially enter corresponding registers for comparing sizes, each clock generates the maximum value at the same moment, and compared with a current pixel value, a filter result is output. Robinson filtering achieved by the invention is in horizontal and vertical directions, the size is 7 by 7 pixels, and a template is 24 pixels at an edge; and all hardwares work in a pixel clock synchronous state, primary comparison of two register numerical values is only generated in each clock period, and clock frequency is favorably improved. The Robinson filtering design method has the characteristics of interface simplicity, real-time performance and universality and is suitably used for achieving a Robinson filter function in the FPGA.
Description
Technical field
The invention belongs to electronic design technology, is related to the method for designing that a kind of Robinson based on FPGA is filtered.
Background technology
Robinson filtering is a kind of non-linear non-parameter type wave filter, also known as protection band filtering.Its formula is:
fRobinson(i, j) denotation coordination (i, j) pixel Robinson filter result, fpre(i, j) denotation coordination (i, j)
Pixel initial data, ZiRepresent each pixel data of the template chosen.
Robinson filtering is usually used in the detection of Small object, it may be determined that the protection band of a size, internal information retain,
Only filtering is compared with the pixel at edge.
In prior art, Robinson filtering belongs to theoretical research, although has been reported that and can realize in FPGA, but not
See in open source literature, the practical situation of Robinson filtering can not be retrieved.
The content of the invention
Technical problem to be solved
In place of the deficiencies in the prior art, the present invention proposes the design that a kind of Robinson based on FPGA is filtered
Method
Technical scheme
The method for designing that a kind of Robinson based on FPGA is filtered, it is characterised in that step is as follows:
Step 1:6 FIFO memories are set up, input data writes the 1st FIFO, read from the 1st FIFO and write the again
2 FIFO, read from the 2nd FIFO and write the 3rd FIFO again, and the rest may be inferred, until reading from the 6th FIFO;
Step 2:The same column data of adjacent 7 row of data formation image that each clock input data and 6 FIFO read, 7
Individual data sequentially enter correspondence depositor and compare size;
Step 3:Each clock produces 9 maximums, correspond to respectively the 1st row, the 7th row, the 1st row, the 2nd row, the 3rd row, the 4th
Row, the 5th row, the 6th row, the 7th row;
Step 4:1st row of each clock generation synchronization, the 7th row, the 1st row, the maximum of the 7th row, with current picture
Plain value compares, and exports filter result.
The Robinson is filtered into horizontal vertical direction, and size is 7 × 7 pixels, and template is 24 pixels at edge.
Beneficial effect
The method for designing that a kind of Robinson based on FPGA proposed by the present invention is filtered, using 6 FIFO memories according to
Secondary connection, input data indatabus are read from 6 FIFO according to this, the data that each clock input data and 6 FIFO read
The same column data of adjacent 7 row of image is formed, 7 data sequentially enter correspondence depositor and compare size, and each clock produces same
The maximum at moment, is compared with current pixel value, exports filter result.The Robinson that the present invention is realized is filtered into level and hangs down
Nogata is to size is 7 × 7 pixels, and template is 24 pixels at edge.Synchronous regime of whole hardware efforts in pixel clock,
The one-level that each clock cycle only has 2 register values compares, and is conducive to improving clock frequency.
Interface of the present invention is simple, the characteristics of with real-time, versatility, it is adaptable to realize that in FPGA Robinson is filtered
Function.
Description of the drawings
Fig. 1 is the main hardware structure schematic diagram of the present invention;
Fig. 2 is the corresponding relation figure of maximum value register and pixel.
Specific embodiment
In conjunction with embodiment, accompanying drawing, the invention will be further described:
The embodiment of the present invention:The Robinson of realization is filtered into horizontal vertical direction, and size is 7 × 7 pixels, and template is side
24 pixels of edge.
Synchronous regime of whole hardware efforts in pixel clock, each clock cycle only have the one-level of 2 register values
Relatively.This hardware designs are conducive to being operated under higher clock frequency.
6 FIFO memories are sequentially connected, and input data indatabus writes the 1st FIFO, read from the 1st FIFO
The 2nd FIFO is write again, is read from the 2nd FIFO and is write the 3rd FIFO again, the rest may be inferred, until reading from the 6th FIFO.
The same column data of adjacent 7 row of data formation image that each clock input data and 6 FIFO read, 7 data
Sequentially enter correspondence depositor and compare size.Each clock produces 9 maximums, correspond to respectively the 1st row, the 7th row, the 1st row, the
2 row, the 3rd row, the 4th row, the 5th row, the 6th row, the 7th row.Each clock produce the 1st row of synchronization, the 7th row, the 1st row, the 7th
The maximum of row, is compared with current pixel value, exports filter result.
Embodiment is designed using VHDL program:
VHDL program
This program is compiled in Xilinx ISE 13.2 and is passed through.
Claims (2)
1. the method for designing that a kind of Robinson based on FPGA is filtered, it is characterised in that step is as follows:
Step 1:6 FIFO memories are set up, input data writes the 1st FIFO, read from the 1st FIFO and write the 2nd again
FIFO, reads from the 2nd FIFO and writes the 3rd FIFO again, and the rest may be inferred, until reading from the 6th FIFO;
Step 2:The same column data of adjacent 7 row of data formation image that each clock input data and 6 FIFO read, 7 numbers
Compare size according to correspondence depositor is sequentially entered;
Step 3:Each clock produces 9 maximums, correspond to respectively the 1st row, the 7th row, the 1st row, the 2nd row, the 3rd row, the 4th row,
5th row, the 6th row, the 7th row;
Step 4:1st row of each clock generation synchronization, the 7th row, the 1st row, the maximum of the 7th row, with current pixel value
Relatively, export filter result.
2. according to claim 1, it is characterised in that:The Robinson is filtered into horizontal vertical direction, and size is 7 × 7
Pixel, 24 pixels of the template for edge.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610919283.9A CN106570230A (en) | 2016-10-21 | 2016-10-21 | Robinson filtering design method based on FPGA (Field Programmable Gate Array) |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610919283.9A CN106570230A (en) | 2016-10-21 | 2016-10-21 | Robinson filtering design method based on FPGA (Field Programmable Gate Array) |
Publications (1)
Publication Number | Publication Date |
---|---|
CN106570230A true CN106570230A (en) | 2017-04-19 |
Family
ID=58534146
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201610919283.9A Pending CN106570230A (en) | 2016-10-21 | 2016-10-21 | Robinson filtering design method based on FPGA (Field Programmable Gate Array) |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN106570230A (en) |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US7102599B2 (en) * | 2001-09-07 | 2006-09-05 | Pioneer Corporation | Identification method for generated position of dynamic false contour, processing method for image signal, and processing apparatus for image signal |
CN104240240A (en) * | 2014-09-04 | 2014-12-24 | 南京理工大学 | Infrared small target detection method and system based on FPGA |
-
2016
- 2016-10-21 CN CN201610919283.9A patent/CN106570230A/en active Pending
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US7102599B2 (en) * | 2001-09-07 | 2006-09-05 | Pioneer Corporation | Identification method for generated position of dynamic false contour, processing method for image signal, and processing apparatus for image signal |
CN104240240A (en) * | 2014-09-04 | 2014-12-24 | 南京理工大学 | Infrared small target detection method and system based on FPGA |
Non-Patent Citations (7)
Title |
---|
周园园等: "基于5x5模板的红外小目标图像检测算法", 《电子科技》 * |
奚云: "空中红外微弱小目标的检测算法及硬件实现", 《红外光电技术》 * |
李丹: "一种基于自适应背景划分的点目标检测算法", 《红外仿真》 * |
郑文龙等: "复杂背景下红外小目标实时识别方法研究", 《红外与毫米波学报》 * |
陈喆等: "基于Robinson双模板匹配及融合的点目标检测算法", 《红外技术》 * |
马庆军等: "基于FPGA的点目标滤波检测算法", 《光学与光电技术》 * |
黄新栋等: "多波段线扫红外图像的实时预处理系统", 《激光与红外》 * |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN105208275B (en) | A kind of system for supporting to handle in real time in flow data piece | |
CN104104888A (en) | Parallel multi-core FPGA digital image real-time zooming processing method and device | |
CN103761707B (en) | A kind of mean filter method rapidly and efficiently removing image salt-pepper noise | |
CN105025264A (en) | FPGA and USB2.0-based portable computer video monitoring system | |
CN104376534A (en) | Low-power-consumption real-time noise-reduction and sharpening merged preprocessing algorithm for CMOS image sensor | |
CN105048994A (en) | Distributed algorithm applied to FIR filters | |
CN105023250B (en) | A kind of realtime graphic system for adaptive enhancement and method based on FPGA | |
CN103530856B (en) | Method and system for removing salt and pepper noise from Bayer image | |
CN106570230A (en) | Robinson filtering design method based on FPGA (Field Programmable Gate Array) | |
CN103400153A (en) | Serial filtering matching method and system for real-time image identification | |
CN104537610A (en) | Super-resolution image reconstruction method based on Sparse representation and UV channel processing | |
CN101951522A (en) | Vector median filtering implementation method for eliminating pulse noises in color image | |
CN106791499A (en) | A kind of method for realizing improving electron multiplication CCD camera output image signal to noise ratio | |
CN101789120B (en) | Image interpolation method based on cosine polynomial | |
CN106558029A (en) | A kind of image filtering method and device | |
Wei et al. | FPGA Implementation of Hardware Accelerator for Real-time Video Image Edge Detection | |
CN104318534B (en) | A kind of Real-time Two-dimensional convolutional digital filtering system | |
CN102129667B (en) | Image scaling method | |
Qin et al. | Real-Time Image Filtering and Edge Detection Method Based on FPGA | |
CN106339988A (en) | Image defogging method and device | |
CN109873954B (en) | Method for realizing color recovery of Bayer array based on FPGA | |
US8824804B2 (en) | Image processing apparatus and image processing method | |
JP6747305B2 (en) | Data processing device and neural network system | |
Jin et al. | A 12-bit 4928× 3264 pixel CMOS image signal processor for digital still cameras | |
CN204145649U (en) | A kind of video signal collection and synthesis system |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20170419 |