RGB Resampler IP核的测试

关于RGB Resampler IP核的测试

1.RGB Resampler功能描述

将输入的RGB数据流转换成其它格式的RGB数据流。

2.功能验证

设置源图像像素数据为：3X4格式。

设置RGB Resampler参数如下图所示，将24-bit RGB格式转换为40-bit RGBA格式。

顶层文件的编写：

 module top(
     clk_clk,
     reset_reset_n,
     video_rgb_resampler_avalon_rgb_source_ready,
     video_rgb_resampler_avalon_rgb_source_startofpacket,
     video_rgb_resampler_avalon_rgb_source_endofpacket,
     video_rgb_resampler_avalon_rgb_source_valid,
     video_rgb_resampler_avalon_rgb_source_data,
 );

     input clk_clk;
     input reset_reset_n;
     input video_rgb_resampler_avalon_rgb_source_ready;
     output video_rgb_resampler_avalon_rgb_source_startofpacket;
     output video_rgb_resampler_avalon_rgb_source_endofpacket;
     output video_rgb_resampler_avalon_rgb_source_valid;
     output [:]video_rgb_resampler_avalon_rgb_source_data;

  RGB_format u1 (
   .clk_clk                                             (clk_clk),
   .reset_reset_n                                       (reset_reset_n),
   .video_rgb_resampler_avalon_rgb_source_ready         (video_rgb_resampler_avalon_rgb_source_ready),
   .video_rgb_resampler_avalon_rgb_source_startofpacket (video_rgb_resampler_avalon_rgb_source_startofpacket),
   .video_rgb_resampler_avalon_rgb_source_endofpacket   (video_rgb_resampler_avalon_rgb_source_endofpacket),
   .video_rgb_resampler_avalon_rgb_source_valid         (video_rgb_resampler_avalon_rgb_source_valid),
   .video_rgb_resampler_avalon_rgb_source_data          (video_rgb_resampler_avalon_rgb_source_data)
  );

 endmodule 

Testbench的编写：

initial
begin
   reset_reset_n = ; #; reset_reset_n = ;
    clk_clk = ; forever # clk_clk = ~clk_clk;
end

always@(posedge clk_clk or negedge reset_reset_n)
    if(!reset_reset_n)
        video_rgb_resampler_avalon_rgb_source_ready = 'b0;
    else
        video_rgb_resampler_avalon_rgb_source_ready = 'b1;

ModelSIM仿真波形：

波形分析： 输出source_data为40-bit,在start---end之间输出12个数据。

3.将224X224 24-bit RGB图像转换为40-bit RGBA 图像

参数配置：

modelsim仿真并导出modelsim中生成的图像像素数据：

integer w_file;
initial w_file = $fopen("video_rgb_resampler_avalon_rgb_source_data.txt");
always@(posedge clk_clk)
    $fdisplay(w_file, "%d", video_rgb_resampler_avalon_rgb_source_data); 

40-bit RGBA数据格式：

在Matlab中进行数据处理：

fd = fopen('video_rgb_resampler_avalon_rgb_source_data.txt');
A =uint64(fscanf(fd,'%f'));
fclose(fd);
B = uint64(zeros(224,224));

A1 = uint64(zeros(50176,1));
A2 = uint64(zeros(50176,1));
A3 = uint64(zeros(224,224));

R1 = uint64(zeros(50176,1));
R2 = uint64(zeros(50176,1));
R3 = uint64(zeros(224,224));

G1 = uint64(zeros(50176,1));
G2 = uint64(zeros(50176,1));
G3 = uint64(zeros(224,224));

B1 = uint64(zeros(50176,1));
B2 = uint64(zeros(224,224));

for j = 1:224
    for i = 1:224
        B(j,i) = A((j-1)*224+i);
    end
end

% A---1111 1111 11 --- 0000 0000 00 --- 0000 0000 00 --- 0000 0000 00
% R---0000 0000 00 --- 1111 1111 11 --- 0000 0000 00 --- 0000 0000 00
% G---0000 0000 00 --- 0000 0000 00 --- 1111 1111 11 --- 0000 0000 00
% B---0000 0000 00 --- 0000 0000 00 --- 0000 0000 00 --- 1111 1111 11  

a = uint64(bin2dec('1111111111000000000000000000000000000000'));
r = uint64(bin2dec('0000000000111111111100000000000000000000'));
g = uint64(bin2dec('0000000000000000000011111111110000000000'));
b = uint64(bin2dec('0000000000000000000000000000001111111111'));

%---A
for i = 1:50176
    A1(i) = bitand(A(i),a);
end

for i = 1:50176
    A2(i) = bitshift(A1(i),-30);
end

for j = 1:224
    for i = 1:224
        A3(j,i) = A2((j-1)*224+i);
    end
end

%---R
for i = 1:50176
    R1(i) = bitand(A(i),r);
end

for i = 1:50176
    R2(i) = bitshift(R1(i),-20);
end

for j = 1:224
    for i = 1:224
        R3(j,i) = R2((j-1)*224+i);
    end
end

%---G
for i = 1:50176
    G1(i) = bitand(A(i),g);
end

for i = 1:50176
    G2(i) = bitshift(G1(i),-10);
end

for j = 1:224
    for i = 1:224
        G3(j,i) = G2((j-1)*224+i);
    end
end

%---B
for i = 1:50176
    B1(i) = bitand(A(i),b);
end

for j = 1:224
    for i = 1:224
        B2(j,i) = B1((j-1)*224+i);
    end
end

需要注意的问题：

源文件中的数据值比较大，如果以十进制的方式读入数据，会出现如下问题：

如果以16进制的形式读入，则不会出现此问题：