Shulou(Shulou.com)06/01 Report--

What is the simulation implementation of face location based on FPGA? I believe many inexperienced people are at a loss about it. Therefore, this paper summarizes the causes and solutions of the problem. Through this article, I hope you can solve this problem.

Simulation realization of face location based on FPGA

1 Overview

In the skin color recognition algorithm, the commonly used color space is Ycbcr,Y for luminance, cb for blue component and cr for red component.

Because the skin color is less affected by the luminance information in the YCbCr space, the algorithm directly considers the CbCr component of the YCbCr space and maps it to a two-dimensional independently distributed CbCr space. In CbCr space, the clustering of skin color is good, and the artificial threshold method is used to separate skin color from non-skin color regions to form a binary image.

The formula for converting RGB to YCbCr is:

Y = 0.257*R+0.564*G+0.098*B+16

Cb=-0.148*R-0.291*G+0.439*B+128

Cr = 0.439*R-0.368*G-0.071*B+128

The conditions for determining skin color are often used as follows:

Cb > 77 years & Cb

133 years & Cr < 173

The goal of this section is to recognize the skin color of a person's face in a complex environment, extract the skin color, and complete the location of the face.

2 experimental procedure

Fig. 1 original diagram of the experiment

Since figure 1 is not an image of 800mm 600, we use a drawing tool to convert figure 1 into 800mm 600.

Use matlab to find the range of human face skin color as shown in figure 2.

Figure 2 matlab finds the skin color range of a human face (cb,cr)

The face color is converted to black, and other colors are converted to white background.

/ *

Modulename: face_location.v

Description: face location

Data: 2018-11-21

E-mail: 137194782@qq.com

Official account of Wechat: FPGA Open Source Studio

, /

`timescale1ns/1ps

Moduleface_location (

Input pix_clk

Input reset_n

Input [23:0] i_rgb

Input [23:0] i_gray

Input [23:0] i_ycbcr

Input i_h_sync

Input i_v_sync

Input i_de

Output [23:0] skin_binary_image

Output [23:0] skin_rgb_image

Output [23:0] skin_gray_image

Output o_h_sync

Output o_v_sync

Output o_de

);

/ / *

/ / skindetection parameter define

/ / *

ParameterY_LOW = 8'd35

ParameterY_HIGH = 8'd60

ParameterCB_LOW = 8'd93

ParameterCB_HIGH = 8'd120

ParameterCR_LOW = 8'd135

ParameterCR_HIGH = 8'd165

Wire [7: 0] y_8b

Wire [7: 0] cb_8b

Wire [7: 0] cr_8b

Reg h_sync_delay

Reg v_sync_delay

Reg de_delay

Reg [23:0] skin_binary_r

Reg [23:0] skin_rgb_r

Reg [23:0] skin_gray_r

Assign yearly 8b = i_ycbcr [23:16]

Assign cb_8b = i_ycbcr [15:8]

Assign cr_8b = i_ycbcr [7:0]

Always@ (posedge pix_clk or negedge reset_n) begin

If (! reset_n) begin

Skin_binary_r

Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.