A Circle Detection Method Based on Optimal A Circle Detection Method - - PowerPoint PPT Presentation
A Circle Detection Method Based on Optimal A Circle Detection Method Based on Optimal Parameter Statistics in Embedded Vision Parameter Statistics in Embedded Vision Xiaofeng Lu [1],[2] [1] Shanghai Key Laboratory of Digital Media Processing and
[1] Shanghai Key Laboratory of Digital Media Processing and Transmissions Shanghai Jiao Tong University, Shanghai, China [2] School of Communication and Information Engineering Shanghai University, Shanghai, China
2 2 2 0 1 3 年1 0 月2 2 日
In this paper, we propose a circle detection method based on the
It employs fast median filtering based Canny edge detection algorithm (FMFCanny) to obtain edge information And real-time three points determination based on optimal parameters statistics circle detection is implemented in FPGA circuit The pipeline processing of FIFO and parallelize operation of registers in FPGA can detects single circle in videos accurately and robustly
3 3 2 0 1 3 年1 0 月2 2 日
Proposed algorithm consists of two parts in FPGA circuits as shown in Fig.1. First, circle edge image is obtained via FMFCanny module Then, circle optimal parameters of horizontal coordinate a, vertical coordinate b and circle’s radius square r2 are calculated through OPSCD module based on the input video and circle edge image
Fig.1. The algorithm’s implementation FPGA architecture
4 4 2 0 1 3 年1 0 月2 2 日
We improve the classic Canny operator in two aspects: (1) Use median filtering instead of Gaussian filtering to suppress grain noise in the image, while edge information is well preserved (2) In non-maximum suppression, only when the reference point’s gradient value is greater than four neighboring points in the gradient direction, it can be viewed as edge point. Since it is time consuming and causes a waste of resources to computes inverse trigonometric functions in FPGA, we divide the gradient direction into eight regions in 360o. The pixel will be treated as an edge point, when its G(x, y) is greater than G(x+1, y-1), G(x, y- 1), G(x, y+1) and G(x-1, y+1).
Fig.2. Flow chart of FMFCanny
5 5 2 0 1 3 年1 0 月2 2 日
Fast median filtering with parallel processing capabilities in FPGA Fast parallel median filtering is achieved by 3×3 template Fast parallel median filtering in FPGA caches two FIFOs and one register to get the 3×3 pixel array to get median value of these 9 pixels as Fig.3 shown.
Fig.3. Schematic diagram of fast median filtering
6 6 2 0 1 3 年1 0 月2 2 日
Three points on a circle can determine the circle Assume that the coordinates of the three points are (x1, y1), (x2, y2) and (x3, y3), the center’s horizontal and vertical coordinates of a, b and radius r, can be calculated by formula (1)(2)(3) )) )( ( ) )( (( 4 ) ( 2 ) ( ) ( 2 ) (
1 2 1 3 1 3 1 2 1 3 2 1 2 1 2 3 2 3 1 2 2 1 2 1 2 2 2 2
y y x x y y x x y y y x y x y y y x y x a − − − − − − × + − + − × + − + =
(1)
)) )( ( ) )( (( 4 ) ( ) ( 2 ) ( ) ( 2
1 2 1 3 1 3 1 2 2 1 2 1 2 3 2 3 1 3 2 1 2 1 2 2 2 2 1 2
y y x x y y x x y x y x x x y x y x x x b − − − − − + − + − × + − + − × =
(2)
3 , 2 , 1 ) ( ) (
2 2 2
= − + − = i b y a x r
i i
(3)
7 7 2 0 1 3 年1 0 月2 2 日
Statistical accumulation is presented to choose the largest number of calculated values as the optimal parameters instead of the average
parameter statistics consists of three modules
Fig.4. Circle detection algorithm based on optimal parameter statistics
8 8 2 0 1 3 年1 0 月2 2 日
The optimal parameter statistices is the key module of our method 2048 RAMs’ addresses represent the circle parameters values RAM address is initialized to 0, and accumulate when the same parameter detected While negative pulse of horizontal sync signal appears, RAMs which have maximum accumulated values will output their addresses as horizontal coordinate a, vertical coordinate b, and radius r as the
2 0 1 3 年1 0 月2 2 日
Fig.5. Key operation of optimal parameter statistics
2 i
r
2
9 9
FPGA hardware based embedded vision system as shown in Fig.6 Camera resolution is 1300x1024 at 43MHz pixel frequency and captures real-time black and white video It takes 0.24 ns to calculate edge information and 17 ms to compute circle optimal parameters, which ensure the real-time processing
Fig.6. FPGA based embedded visual platform
10 10 2 0 1 3 年1 0 月2 2 日
We get circle images to do edge detection through classic Canny and proposed FMFCanny respectively, as shown in Fig.7 In Fig.7 (b), there is edge loss in vertical and horizontal directions. In comparison, the edge curve is more continuous and clear without edge loss in Fig.7 (c)
Fig.7. (a) original image, (b) result of [1], (c) result of FMFCanny
11 11 2 0 1 3 年1 0 月2 2 日
Due to optimal parameter statistics, the algorithm gets good results in a simple background, without error caused by circle parameter accumulation and average operations
Fig.8. (a)(d) original images, (b)(e) results of CGM[4], (c)(f) results of proposed OPSCD
12 12 2 0 1 3 年1 0 月2 2 日
Signal Tap II is used to record real-time circle parameters. There are 20 times computation records as Fig. 9 shown.
2 0 1 3 年1 0 月2 2 日 13 13
Fig.9. Comparison between the two circle detection algorithm with actual values in 20 times computation records
(b) (c)
g represents circle
center’s horizontal coordinate, vertical coordinate and radius square in CGM, and ao, bo, r2
horizontal coordinate, vertical coordinate and radius square in OPSCD
According to the analysis of circle parameters, the error ranges of the three circle parameters (a, b, r2) in CGM are [0 0.59%], [0 0.95%], [0 2.24%], while in our proposed method the corresponding error ranges are [0 0.45%], [0 0.48%], [0 0.96%] This article puts forward a FPGA based embedded vision platform, and proposes a circle detection algorithm based on optimal parameter statistics With the module of optimal parameter statistics, it reduces the computational complexity and ensures the accuracy and robustness
14 14 2 0 1 3 年1 0 月2 2 日