v4l2 spec 中文 Ch02

Chapter 2.  图像格式(Image Formats)

v4l2 全五章下载：

v4l2中文手册 ch02

V4L2 API主要是设计用于设备和应用程序进行图片数据交换的。结构v4l2_pix_format定义了一张图片在内存中的格式和布局。格式是通过VIDIOC_S_FMT协商得到的。（这里涉及的重点是关乎视频采集和输出，对于overlay帧缓冲格式，您还是看看VIDIOC_G_FBUF)。

 

Table 2-1. structv4l2_pix_format

__u32                                   width                               Image width in pixels.

__u32                                    height                              Image height inpixels.

应用程序设置这些域是用来请求图片尺寸，驱动会返回最接近的合适值。在平面格式情形中…（不知是不是我下规范不好怎么地，在这种列表式中总会有些地方不完整，当然我看过了，这不太影响我们的工作）

Applicationsset these fields to request an image size, drivers return the closest possiblevalues. In case of planar formats

__u32                                   pixelformat                   The pixel format or type of compression, set by

theapplication. This is a little endian four

charactercode. V4L2 defines standard RGB

formatsin Table 2-1, YUV formats in Section 2.5,

andreserved codes in Table 2-8

enum v4l2_field                 field                                Video images are typically interlaced.

                                                                                    Applicationscan request to capture or output only

                                                                                    thetop or bottom field, or both fields interlaced or

                                                                                    sequentiallystored in one buffer or alternating in

                                                                                    separatebuffers. Drivers return the actual field

                                                                                    orderselected. For details see Section 3.6.

__u32                                   bytesperline                  Distance in bytes between theleftmost pixels in

                                                                                    twoadjacent lines.

应用程序和驱动都可以设置这些域以请求在每行的末尾的填补位。然而驱动却可能予以忽略…

Both applications and drivers can set this field torequest padding bytes at the end of each line. Drivers however may ignore

__u32                                   sizeimage                     Size in bytes of the buffer to hold a complete

                                                                                    image,set by the driver. Usually this is

                                                                                    bytesperlinetimes height. When the image

                                                                                    consistsof variable length compressed data this is

                                                                                    themaximum number of bytes required to hold an

                                                                                    image.

enum v4l2_colorspace     colorspace                    Thisinformation supplements the pixelformat

                                                                                        andmust be set by the driver,see Section2.2.

__u32                                   priv                                 Reserved for custom (driver defined)additional

                                                                                        informationabout formats. When not used drivers

                                                                                        andapplications must set this field to zero.

2.1.  标准图形格式(Standard Image Formats)

    为了在驱动和应用程序间进行图片交换，就必须有一个双方都能辨析（统一的）的图片数据格式。V4L2包含了诸多格式，而本节就来谈谈V4L2规范中的标准图形格式。

    当然，V4L2驱动并不局限于这些个格式。相反，可以有定制的格式。在这种情况下，当需要时应用程序就可能依赖解码器来将这种格式转换成标准类型的。但是数据依然可以以专有（定制的）格式来进行数据的存储和重新获取。例子说，假设一个设备支持专有的压缩格式，那么，应用程序仍然可以以这种格式进行采集和保存数据，已达到节省空间的目的。而当要在X Windows端显示时，就可以用解码器进行解码并显示。

    最后尽管如此，我们还是需要一些个标准格式的。只有定义了明确的标准格式，才可以保证V4L2规范的完整性嘛。

    （接下来的这一段主要说标准格式，是未压缩的，其在内存的布局是怎么样的，就不译了）

    在V4L2中，每种格式都个标识符形如PIX_FMT_XXX，定义在videodev.h头文件中。这些标识符代表了在下面列出的4字节代码，当然了他们和Windows（相对的是Linux或者Unix）世界里的用法是不一样的。

2.2.  色彩空间(Colorspaces)

GammaCorrection

[to do]

E’R = f(R)

         E’G =f(G)

         E’B =f(B)

Constructionof luminance and color-difference signals

[to do]

E’_Y = Coeff_R E’_R+ Coeff_G E’_G + Coeff_B E’_B

(E’_R - E’_Y) =E’_R - Coeff_R E’_R - Coeff_G E’_G- Coeff_B E’_B

(E’_B - E’_Y) =E’_B - Coeff_R E’_R - Coeff_G E’_G- Coeff_B E’_B

Re-normalizedcolor-difference signals

The color-difference signals arescaled back to unity range [-0.5;+0.5]:

K_B = 0.5 / (1 - Coeff_B)

K_R = 0.5 / (1 - Coeff_R)

P_B = K_B (E’_B- E’_Y) = 0.5 (Coeff_R / Coeff_B) E’_R+ 0.5 (Coeff_G / Coeff_B) E’_G + 0.5 E’_B

P_R = K_R (E’_R- E’_Y) = 0.5 E’_R + 0.5 (Coeff_G / Coeff_R)E’_G + 0.5 (Coeff_B/ Coeff_R) E’_B

Quantization

[to do]

Y’ = (Lum. Levels - 1) · E’_Y+ Lum. Offset

C_B = (Chrom. Levels - 1) ·P_B + Chrom. Offset

C_R =(Chrom. Levels - 1) · P_R + Chrom. Offset

给大伙提供个介绍RGB和YUV关系的文档（网上找的，在此感谢原作者！）

JPEG 简易文档 V1.0 - GameRes.com.htm(如果打不开直接去搜一下吧)

 

Example 2-1. ITU-R Rec.BT.601 color conversion

ForwardTransformation

intER, EG, EB;           /* gamma correctedRGB input [0;255] */

intY1, Cb, Cr;            /* output [0;255]*/

 

doubler, g, b;             /* temporaries */

doubley1, pb, pr;

 

int

clamp (double x)

{

int r = x; /* round to nearest */

 

if (r < 0)                        return 0;

else if (r > 255)           return 255;

else                              return r;

}

 

r= ER / 255.0;

g= EG / 255.0;

b= EB / 255.0;

 

y1= 0.299 * r + 0.587 * g + 0.114 * b;

pb= -0.169 * r - 0.331 * g + 0.5 * b;

pr= 0.5 * r - 0.419 * g - 0.081 * b;

 

Y1= clamp (219 * y1 + 16);

Cb= clamp (224 * pb + 128);

Cr= clamp (224 * pr + 128);

 

/*or shorter */

y1= 0.299 * ER + 0.587 * EG + 0.114 * EB;

 

Y1= clamp ( (219 / 255.0) * y1 + 16);

Cb= clamp (((224 / 255.0) / (2 - 2 * 0.114)) * (EB - y1) + 128);

Cr= clamp (((224 / 255.0) / (2 - 2 * 0.299)) * (ER - y1) + 128);

 

Inverse Transformation

int Y1, Cb, Cr;                      /* gamma pre-correctedinput [0;255] */

int ER, EG, EB;                    /* output [0;255] */

 

double r, g, b;                       /* temporaries */

doubley1, pb, pr;

 

int

clamp(double x)

{

int r = x; /* round to nearest */

if (r < 0)                        return 0;

else if (r > 255)          return 255;

else                             return r;

}

 

y1= (255 / 219.0) * (Y1 - 16);

pb= (255 / 224.0) * (Cb - 128);

pr= (255 / 224.0) * (Cr - 128);

 

r= 1.0 * y1 + 0 * pb + 1.402 * pr;

g= 1.0 * y1 - 0.344 * pb - 0.714 * pr;

b= 1.0 * y1 + 1.772 * pb + 0 * pr;

 

ER= clamp (r * 255); /* [ok? one should prob. limit y1,pb,pr] */

EG= clamp (g * 255);

EB= clamp (b * 255);

 

Table 2-2. enum v4l2_colorspace

(详见P₄₂)

2.3.  索引格式(Indexed Format)

   在这种格式中，每个像素点都用一个8bit来表示，用来索引256色的调色板。这是特地为视频输出overlay准备的。没有用来访问该调色板的ioctl方法，但可以且仅可以通过Linux的帧缓冲API（头一回接触）来进行访问。

Table2-3. Indexed Image Format(P₄₂-P₄₃)

2.4.  RGB格式(RGB Formats)

Packed RGB Formats

       Name

         Packed RGB formats  – Packed RGB formats

 

       Description

这些格式定义用来匹配传统PC的图形帧缓冲的像素格式的。每像素占据8,16，24或者32个bit。这就是填充像素(packed-pixel)格式,意味着每个像素点在内存中是一个挨着一个的。

当使用这些格式中的某一个时，驱动应该报告给色彩空间：V4L2_COLORSPACE_SRGB。

 

 

 

 

Table2-1. Packed RGB Image Formats

Identifier Code	Byte 0 7654/3210	Byte 1	Byte 2	Byte 3
V4L2_PIX_FMT_RGB332 ‘RGB1’	b1 b0 g2 g1 g0 r2 r1 r0
V4L2_PIX_FMT_RGB444 ‘R444’	g3 g2 g1 g0 b3 b2 b1 b0	a3 a2 a1 a0 r3 r2 r1 r0
V4L2_PIX_FMT_RGB555 ‘RGBO’	g2 g1 g0 r4 r3 r2 r1 r0	a b4 b3 b2 b1 b0 g4 g3
V4L2_PIX_FMT_RGB555X ‘RGBQ’	a b4 b3 b2 b1 b0 g4 g3	g2 g1 g0 r4 r3 r2 r1 r0
V4L2_PIX_FMT_RGB565 ‘RGBP’	g2 g1 g0 r4 r3 r2 r1 r0	b4 b3 b2 b1 b0 g5 g4 g3
V4L2_PIX_FMT_RGB565X ‘RGBR’	b4 b3 b2 b1 b0 g5 g4 g3	g2 g1 g0 r4 r3 r2 r1 r0
V4L2_PIX_FMT_BGR24 ‘BGR3’	b7 b6 b5 b4 b3 b2 b1 b0	g7 g6 g5 g4 g3 g2 g1 g0	r7 r6 r5 r4 r3 r2 r1 r0
V4L2_PIX_FMT_RGB24 ‘RGB3’	r7 r6 r5 r4 r3 r2 r1 r0	g7 g6 g5 g4 g3 g2 g1 g0	b7 b6 b5 b4 b3 b2 b1 b0
V4L2_PIX_FMT_BGR32 ‘BGR4’	b7 b6 b5 b4 b3 b2 b1 b0	g7 g6 g5 g4 g3 g2 g1 g0	r7 r6 r5 r4 r3 r2 r1 r0	a7 a6 a5 a4 a3 a2 a1 a0
V4L2_PIX_FMT_RGB32 ‘RGB4’	r7 r6 r5 r4 r3 r2 r1 r0	g7 g6 g5 g4 g3 g2 g1 g0	b7 b6 b5 b4 b3 b2 b1 b0	a7 a6 a5 a4 a3 a2 a1 a0

一个驱动到底支持哪种RGB格式，你可以去LinuxTV v4l-dvb代码库下个工具测试下。可以访问http://linuxtv.org/repo/ 来获得更多信息。

 

V4L2_PIX_FMT_SBGGR8 (’BA81’)

Name

         V4L2_PIX_FMT_SBGG8 --- Bayer RGB format

       Description

这通常是数字摄像机的本征（native）格式，用来表明CCD设备上传感器的布局的。每个像素点就一个红色或者绿色或者蓝色。缺失部分必须用邻近像素来内插补充。从左到右，第一行包含一个blue和green，第二行是一个green和red。这种组合会向右向下每隔一行一列进行排布。

Example 2-1. V4L2_PIX_FMT_SBGGR8 4 × 4 pixel image

Byte Order. Each cell is one bytew.

start + 0:            B₀₀             G₀₁            B₀₂             G₀₃

start + 4:            G₁₀            R₁₁            G₁₂            R₁₃

start + 8:            B₂₀             G₂₁            B₂₂             G₂₃

start + 12:          G₃₀           R₃₁            G₃₂            R₃₃

 

V4L2_PIX_FMT_SBGGR16 (’BA82’)

       Name

         V4L2_PIX_FMT_SBGGR16— Bayer RGB format

Description

                                                                                     Tobe continued…