服务器之家

服务器之家 > 正文

C++ HLSL实现简单的图像处理功能

时间:2021-04-30 15:41     来源/作者:三米上

由于对于dxva2解码得到的数据不宜copy回内存给CPU处理,所以最好的办法是在GPU上直接进行处理。D3D的像素着色器能够对像素直接进行操作,实现点运算极其简单方便,简单的卷积运算效果也非常好。但D3D9的限制也很多,对于过于复杂的图像处理则显得有些不能胜任。

1.点运算

点运算用HLSL非常容易实现,几乎是公式怎么写,代码就怎么写。以RGB转灰度图显示为例:

?
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
texture Tex0 ;
int iFlag = 0 ;
float aValue= 0.0 ;
float bValue= 0.0 ;
sampler2D YTex =
sampler_state
{
  Texture = <Tex0> ;
  MipFilter = LINEAR ;
  MinFilter = LINEAR ;
  MagFilter = LINEAR ;
 
  AddressU = CLAMP ;
  AddressV = CLAMP ;
};
struct PS_INPUT
{
  float2 uvCoords0 : TEXCOORD0 ;
};
float4 Main( PS_INPUT input ) : COLOR0
{
  float4 yuvColor ;
  //rgb to gray 不知道是不是这么显示的,姑且这么认为
  float gray = tex2D( YTex, input.uvCoords0 ).r * 0.299 + tex2D( YTex, input.uvCoords0 ).g * 0.587 + tex2D( YTex, input.uvCoords0 ).b * 0.114 ;
  float s = 0 ;
  if(iFlag == 0)
  {
    s = aValue * gray + bValue/255 ;
  }
  else if(iFlag == 1)
  {
    s = aValue * log(1+gray) ;
  }
  else if(iFlag == 2)
  {
    s = aValue * pow(abs(gray),bValue) ;
  }
  yuvColor.r = s ;
  yuvColor.g = s ;
  yuvColor.b = s ;
  yuvColor.a = 1.0 ;
  return yuvColor ;
}

点运算如此简单是因为GPU是并行运算的,我个人认为可以看成是每一个像素点(BGRA)对应一个线程,这大概就是OpenCL中所谓的数据并行。这是一个非常简单的程序,指令数少,程序结构也很简单,shader 的版本用2.0就可以轻松编过。

2.卷积运算举例

指令数较多的情况2.0版本的shader就搞不定了,上3.0版本可以做一些简单的卷积运算。以中值滤波为例:

?
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
texture Tex0 ;
matrix WorldMatrix;
matrix ViewMatrix;
matrix ProjMatrix;
sampler2D YTex =
sampler_state
{
  Texture = <Tex0> ;
  MipFilter = LINEAR ;
  MinFilter = LINEAR ;
  MagFilter = LINEAR ;
  AddressU = CLAMP ;
  AddressV = CLAMP ;
};
struct VS_INPUT
 {
  float4 pos  : POSITION;
  float4 color : COLOR0;
  float2 tex  : TEXCOORD0;
 };
//
struct VS_OUTPUT
 {
  float4 pos   : POSITION;
  float4 color  : COLOR0;
  float2 tex   : TEXCOORD0;
 };
float2 g_v4ScreenSize;
int ksize = 1 ;
float fLeft = -1.0f ;
float fTop = -1.0f ;
float fRight = -1.0f ;
float fBottom = -1.0f ;
//--------------------------------- BurTechnique --------------------------------------
VS_OUTPUT MainVS_Screen( VS_INPUT In )
{
  VS_OUTPUT Out = ( VS_OUTPUT )0;
  float4x4 matWorldView = mul(WorldMatrix,ViewMatrix);
  float4x4 matProject = mul(matWorldView,ProjMatrix);
  Out.pos = mul(In.pos,matProject);
  Out.tex = In.tex;
  Out.color = In.color;
  return Out;
}
float4 MainPS_Screen( VS_INPUT In ) : COLOR0
{
  float4 outColor = tex2D( YTex, In.tex ).rgba ;
  if(ksize <= 1 || ksize%2 == 0)
  {
    return outColor ;
  }
  if( ksize > 11 || ksize < 3)
  {
    return outColor ;
  }
  if(!(In.tex.x < fRight && In.tex.y < fBottom && In.tex.x > fLeft && In.tex.y > fTop))
  {
    return outColor ;
  }
  // 纹理大小
  float2 TexSize = float2( g_v4ScreenSize.x , g_v4ScreenSize.y );
  float x_off = 1.0f / TexSize.x;
  float y_off = 1.0f / TexSize.y;
  float2 fX0Y0 = In.tex - float2(x_off * ksize/2, y_off*ksize/2) ;
  float3 sum = {0.0f, 0.0f, 0.0f} ;
  if(ksize >= 3)
  {
    sum += tex2D( YTex , fX0Y0 + float2(x_off * 0, y_off*0)).rgb;
    sum += tex2D( YTex , fX0Y0 + float2(x_off * 0, y_off*1)).rgb;
    sum += tex2D( YTex , fX0Y0 + float2(x_off * 0, y_off*2)).rgb;
    sum += tex2D( YTex , fX0Y0 + float2(x_off * 1, y_off*0)).rgb;
    sum += tex2D( YTex , fX0Y0 + float2(x_off * 1, y_off*1)).rgb;
    sum += tex2D( YTex , fX0Y0 + float2(x_off * 1, y_off*2)).rgb;
    sum += tex2D( YTex , fX0Y0 + float2(x_off * 2, y_off*0)).rgb;
    sum += tex2D( YTex , fX0Y0 + float2(x_off * 2, y_off*1)).rgb;
    sum += tex2D( YTex , fX0Y0 + float2(x_off * 2, y_off*2)).rgb;
  }
  if(ksize >= 5)
  {
    sum += tex2D( YTex , fX0Y0 + float2(x_off * 3, y_off*0)).rgb;
    sum += tex2D( YTex , fX0Y0 + float2(x_off * 3, y_off*1)).rgb;
    sum += tex2D( YTex , fX0Y0 + float2(x_off * 3, y_off*2)).rgb;
    sum += tex2D( YTex , fX0Y0 + float2(x_off * 3, y_off*3)).rgb;
    sum += tex2D( YTex , fX0Y0 + float2(x_off * 3, y_off*4)).rgb;
    sum += tex2D( YTex , fX0Y0 + float2(x_off * 4, y_off*0)).rgb;
    sum += tex2D( YTex , fX0Y0 + float2(x_off * 4, y_off*1)).rgb;
    sum += tex2D( YTex , fX0Y0 + float2(x_off * 4, y_off*2)).rgb;
    sum += tex2D( YTex , fX0Y0 + float2(x_off * 4, y_off*3)).rgb;
    sum += tex2D( YTex , fX0Y0 + float2(x_off * 4, y_off*4)).rgb;
    sum += tex2D( YTex , fX0Y0 + float2(x_off * 0, y_off*3)).rgb;
    sum += tex2D( YTex , fX0Y0 + float2(x_off * 1, y_off*3)).rgb;
    sum += tex2D( YTex , fX0Y0 + float2(x_off * 2, y_off*3)).rgb;
    sum += tex2D( YTex , fX0Y0 + float2(x_off * 0, y_off*4)).rgb;
    sum += tex2D( YTex , fX0Y0 + float2(x_off * 1, y_off*4)).rgb;
    sum += tex2D( YTex , fX0Y0 + float2(x_off * 2, y_off*4)).rgb;
  }
  if(ksize >= 7)
  {
    sum += tex2D( YTex , fX0Y0 + float2(x_off * 5, y_off*0)).rgb;
    sum += tex2D( YTex , fX0Y0 + float2(x_off * 5, y_off*1)).rgb;
    sum += tex2D( YTex , fX0Y0 + float2(x_off * 5, y_off*2)).rgb;
    sum += tex2D( YTex , fX0Y0 + float2(x_off * 5, y_off*3)).rgb;
    sum += tex2D( YTex , fX0Y0 + float2(x_off * 5, y_off*4)).rgb;
    sum += tex2D( YTex , fX0Y0 + float2(x_off * 5, y_off*5)).rgb;
    sum += tex2D( YTex , fX0Y0 + float2(x_off * 5, y_off*6)).rgb;
    sum += tex2D( YTex , fX0Y0 + float2(x_off * 6, y_off*0)).rgb;
    sum += tex2D( YTex , fX0Y0 + float2(x_off * 6, y_off*1)).rgb;
    sum += tex2D( YTex , fX0Y0 + float2(x_off * 6, y_off*2)).rgb;
    sum += tex2D( YTex , fX0Y0 + float2(x_off * 6, y_off*3)).rgb;
    sum += tex2D( YTex , fX0Y0 + float2(x_off * 6, y_off*4)).rgb;
    sum += tex2D( YTex , fX0Y0 + float2(x_off * 6, y_off*5)).rgb;
    sum += tex2D( YTex , fX0Y0 + float2(x_off * 6, y_off*6)).rgb;
    sum += tex2D( YTex , fX0Y0 + float2(x_off * 0, y_off*5)).rgb;
    sum += tex2D( YTex , fX0Y0 + float2(x_off * 1, y_off*5)).rgb;
    sum += tex2D( YTex , fX0Y0 + float2(x_off * 2, y_off*5)).rgb;
    sum += tex2D( YTex , fX0Y0 + float2(x_off * 3, y_off*5)).rgb;
    sum += tex2D( YTex , fX0Y0 + float2(x_off * 4, y_off*5)).rgb;
    sum += tex2D( YTex , fX0Y0 + float2(x_off * 0, y_off*6)).rgb;
    sum += tex2D( YTex , fX0Y0 + float2(x_off * 1, y_off*6)).rgb;
    sum += tex2D( YTex , fX0Y0 + float2(x_off * 2, y_off*6)).rgb;
    sum += tex2D( YTex , fX0Y0 + float2(x_off * 3, y_off*6)).rgb;
    sum += tex2D( YTex , fX0Y0 + float2(x_off * 4, y_off*6)).rgb;
  }
  if(ksize >= 9)
  {
    sum += tex2D( YTex , fX0Y0 + float2(x_off * 7, y_off*0)).rgb;
    sum += tex2D( YTex , fX0Y0 + float2(x_off * 7, y_off*1)).rgb;
    sum += tex2D( YTex , fX0Y0 + float2(x_off * 7, y_off*2)).rgb;
    sum += tex2D( YTex , fX0Y0 + float2(x_off * 7, y_off*3)).rgb;
    sum += tex2D( YTex , fX0Y0 + float2(x_off * 7, y_off*4)).rgb;
    sum += tex2D( YTex , fX0Y0 + float2(x_off * 7, y_off*5)).rgb;
    sum += tex2D( YTex , fX0Y0 + float2(x_off * 7, y_off*6)).rgb;
    sum += tex2D( YTex , fX0Y0 + float2(x_off * 7, y_off*7)).rgb;
    sum += tex2D( YTex , fX0Y0 + float2(x_off * 7, y_off*8)).rgb;
    sum += tex2D( YTex , fX0Y0 + float2(x_off * 8, y_off*0)).rgb;
    sum += tex2D( YTex , fX0Y0 + float2(x_off * 8, y_off*1)).rgb;
    sum += tex2D( YTex , fX0Y0 + float2(x_off * 8, y_off*2)).rgb;
    sum += tex2D( YTex , fX0Y0 + float2(x_off * 8, y_off*3)).rgb;
    sum += tex2D( YTex , fX0Y0 + float2(x_off * 8, y_off*4)).rgb;
    sum += tex2D( YTex , fX0Y0 + float2(x_off * 8, y_off*5)).rgb;
    sum += tex2D( YTex , fX0Y0 + float2(x_off * 8, y_off*6)).rgb;
    sum += tex2D( YTex , fX0Y0 + float2(x_off * 8, y_off*7)).rgb;
    sum += tex2D( YTex , fX0Y0 + float2(x_off * 8, y_off*8)).rgb;
    sum += tex2D( YTex , fX0Y0 + float2(x_off * 0, y_off*7)).rgb;
    sum += tex2D( YTex , fX0Y0 + float2(x_off * 1, y_off*7)).rgb;
    sum += tex2D( YTex , fX0Y0 + float2(x_off * 2, y_off*7)).rgb;
    sum += tex2D( YTex , fX0Y0 + float2(x_off * 3, y_off*7)).rgb;
    sum += tex2D( YTex , fX0Y0 + float2(x_off * 4, y_off*7)).rgb;
    sum += tex2D( YTex , fX0Y0 + float2(x_off * 5, y_off*7)).rgb;
    sum += tex2D( YTex , fX0Y0 + float2(x_off * 6, y_off*7)).rgb;
    sum += tex2D( YTex , fX0Y0 + float2(x_off * 0, y_off*8)).rgb;
    sum += tex2D( YTex , fX0Y0 + float2(x_off * 1, y_off*8)).rgb;
    sum += tex2D( YTex , fX0Y0 + float2(x_off * 2, y_off*8)).rgb;
    sum += tex2D( YTex , fX0Y0 + float2(x_off * 3, y_off*8)).rgb;
    sum += tex2D( YTex , fX0Y0 + float2(x_off * 4, y_off*8)).rgb;
    sum += tex2D( YTex , fX0Y0 + float2(x_off * 5, y_off*8)).rgb;
    sum += tex2D( YTex , fX0Y0 + float2(x_off * 6, y_off*8)).rgb;
  }
  if(ksize >= 11)
  {
    sum += tex2D( YTex , fX0Y0 + float2(x_off * 9, y_off*0)).rgb;
    sum += tex2D( YTex , fX0Y0 + float2(x_off * 9, y_off*1)).rgb;
    sum += tex2D( YTex , fX0Y0 + float2(x_off * 9, y_off*2)).rgb;
    sum += tex2D( YTex , fX0Y0 + float2(x_off * 9, y_off*3)).rgb;
    sum += tex2D( YTex , fX0Y0 + float2(x_off * 9, y_off*4)).rgb;
    sum += tex2D( YTex , fX0Y0 + float2(x_off * 9, y_off*5)).rgb;
    sum += tex2D( YTex , fX0Y0 + float2(x_off * 9, y_off*6)).rgb;
    sum += tex2D( YTex , fX0Y0 + float2(x_off * 9, y_off*7)).rgb;
    sum += tex2D( YTex , fX0Y0 + float2(x_off * 9, y_off*8)).rgb;
    sum += tex2D( YTex , fX0Y0 + float2(x_off * 9, y_off*9)).rgb;
    sum += tex2D( YTex , fX0Y0 + float2(x_off * 9, y_off*10)).rgb;
    sum += tex2D( YTex , fX0Y0 + float2(x_off * 10, y_off*0)).rgb;
    sum += tex2D( YTex , fX0Y0 + float2(x_off * 10, y_off*1)).rgb;
    sum += tex2D( YTex , fX0Y0 + float2(x_off * 10, y_off*2)).rgb;
    sum += tex2D( YTex , fX0Y0 + float2(x_off * 10, y_off*3)).rgb;
    sum += tex2D( YTex , fX0Y0 + float2(x_off * 10, y_off*4)).rgb;
    sum += tex2D( YTex , fX0Y0 + float2(x_off * 10, y_off*5)).rgb;
    sum += tex2D( YTex , fX0Y0 + float2(x_off * 10, y_off*6)).rgb;
    sum += tex2D( YTex , fX0Y0 + float2(x_off * 10, y_off*7)).rgb;
    sum += tex2D( YTex , fX0Y0 + float2(x_off * 10, y_off*8)).rgb;
    sum += tex2D( YTex , fX0Y0 + float2(x_off * 10, y_off*9)).rgb;
    sum += tex2D( YTex , fX0Y0 + float2(x_off * 10, y_off*10)).rgb;
    sum += tex2D( YTex , fX0Y0 + float2(x_off * 0, y_off*9)).rgb;
    sum += tex2D( YTex , fX0Y0 + float2(x_off * 1, y_off*9)).rgb;
    sum += tex2D( YTex , fX0Y0 + float2(x_off * 2, y_off*9)).rgb;
    sum += tex2D( YTex , fX0Y0 + float2(x_off * 3, y_off*9)).rgb;
    sum += tex2D( YTex , fX0Y0 + float2(x_off * 4, y_off*9)).rgb;
    sum += tex2D( YTex , fX0Y0 + float2(x_off * 5, y_off*9)).rgb;
    sum += tex2D( YTex , fX0Y0 + float2(x_off * 6, y_off*9)).rgb;
    sum += tex2D( YTex , fX0Y0 + float2(x_off * 7, y_off*9)).rgb;
    sum += tex2D( YTex , fX0Y0 + float2(x_off * 8, y_off*9)).rgb;
    sum += tex2D( YTex , fX0Y0 + float2(x_off * 0, y_off*10)).rgb;
    sum += tex2D( YTex , fX0Y0 + float2(x_off * 1, y_off*10)).rgb;
    sum += tex2D( YTex , fX0Y0 + float2(x_off * 2, y_off*10)).rgb;
    sum += tex2D( YTex , fX0Y0 + float2(x_off * 3, y_off*10)).rgb;
    sum += tex2D( YTex , fX0Y0 + float2(x_off * 4, y_off*10)).rgb;
    sum += tex2D( YTex , fX0Y0 + float2(x_off * 5, y_off*10)).rgb;
    sum += tex2D( YTex , fX0Y0 + float2(x_off * 6, y_off*10)).rgb;
    sum += tex2D( YTex , fX0Y0 + float2(x_off * 7, y_off*10)).rgb;
    sum += tex2D( YTex , fX0Y0 + float2(x_off * 8, y_off*10)).rgb;
  }
  outColor = float4(sum/(ksize*ksize),1.0f);
  return outColor ;
}
//--------------------------- 技术---------------------------
technique BurTechnique
{
  pass P0
  
    LightEnable[0] = false;
    VertexShader = compile vs_3_0 MainVS_Screen();
    PixelShader = compile ps_3_0 MainPS_Screen();
  }
}

由于3.0版本的shader似乎不允许pixel shader单独出现,所以我从点运算用像素着色器实现改为用特效来实现。HLSL语法中有if语句,也有for语句,可是这个程序却不厌其烦的把所有的都给列出来来,而没有使用for循环。这是因为在实际使用中发现有一些限制,比如if语句的if(A>B),A与B中必须有一个是常量,就像上面见到的那种形式;for循环中间的判断也是如此,只是在第二层j循环中可以是第一层循环的i,即不可以

?
1
2
3
4
5
6
7
for(int i=0;i<ksize;i++)
{
  for(int j=0;j<ksize1;j++)
    {
  ..........
    }
}

以上代码的ksize与ksize1都必须为常数,例外的情况是ksize1可以为第一层循环的 i 。这个问题不知道后续版本的shader有没有,反正我当前使用的版本有。

另外有一个需要注意的地方是指令数,2.0版本的shader支持的指令数相当少,3.0版本则要多好多,我最长写到了400多条快500条时才导致编译失败。 还有一个需要提醒的是3.0版本的shader只支持D3D 9.0C以后的。如果要求做更为复杂的图像处理,可以的话建议上D3D11,compute shader虽然我没用过,但从介绍来说,应该可以处理一些更为复杂的图像处理。

以上就是本文的全部内容,希望本文的内容对大家的学习或者工作能带来一定的帮助,同时也希望多多支持服务器之家!

原文链接:http://www.cnblogs.com/betterwgo/p/6403522.html

标签:

相关文章

热门资讯

2020微信伤感网名听哭了 让对方看到心疼的伤感网名大全
2020微信伤感网名听哭了 让对方看到心疼的伤感网名大全 2019-12-26
yue是什么意思 网络流行语yue了是什么梗
yue是什么意思 网络流行语yue了是什么梗 2020-10-11
背刺什么意思 网络词语背刺是什么梗
背刺什么意思 网络词语背刺是什么梗 2020-05-22
苹果12mini价格表官网报价 iPhone12mini全版本价格汇总
苹果12mini价格表官网报价 iPhone12mini全版本价格汇总 2020-11-13
Intellij idea2020永久破解,亲测可用!!!
Intellij idea2020永久破解,亲测可用!!! 2020-07-29
返回顶部