Imatest图像质量检测手册

PDF文件

imatest

document

Norman

下载需积分: 38 | 13.15MB | 更新于2024-07-01 | 115 浏览量 | 举报收藏

立即下载

"Imatest说明手册提供了全面的图像质量测试和分析方法，涵盖了从清晰度、主观质量因子(SQF)、噪声、标准化锐化、相机间的锐度比较、色差、镜头眩光到色彩校正矩阵以及ISO感光度和曝光指数等多个关键知识点。手册由Imatest LLC在2009年发布，旨在帮助用户理解和应用Imatest软件进行专业图像质量评估。" 1. 清晰度(Sharpness) - Imatest通过斜边测试来评估图像的清晰度，这涉及到计算细节并提供MTF50结果，帮助用户理解图像在不同频率下的分辨率。 - MTF（Modulation Transfer Function）是衡量光学系统对图像细节保留能力的重要指标。 2. 主观质量因子(SQF) - SQF与MTF相关，但更注重人眼对图像质量的感知。SQF的计算涉及到主观评价和对比度敏感函数(CSF)，提供了一种量化图像主观质量的方法。 3. 噪声(Noise in photographic images) - 噪声会影响图像的清晰度和视觉效果，Imatest探讨了噪声的外观、测量方法和总结，以及与F-stop（光圈大小）的关系。 - 讨论了噪声的数学模型，帮助理解噪声如何随光照条件变化。 4. 标准化锐化(Standardized sharpening) - 在比较不同相机时，标准化锐化是必要的，以消除相机内置处理的影响，确保公平公正的比较。 5. 锐度比较(Sharpness comparisons) - 对多款数码相机的锐度进行了比较，通过结果表格和MTF解读，解析了不同相机在分辨率上的差异。 6. 色差(Chromatic Aberration) - 色差，即颜色 fringing，是光学系统中常见的问题。Imatest介绍了测量色差的方法，以及在去马赛克过程中的影响。 7. 镜头眩光(Veiling glare/Lensflare) - 镜头眩光影响图像的对比度和清晰度，手册描述了测试目标、测量过程以及ISO 9358标准的相关应用。 8. 色彩校正矩阵(Color correction matrix) - 色彩校正矩阵用于校正不同设备间色彩再现的差异，手册涵盖了其数学原理和使用多图表进行色彩校正的方法。 9. ISO感光度和曝光指数(ISO Sensitivity and Exposure Index) - Imatest提供了关于ISO感光度模块的信息，解释了相关的计算公式，并讨论了RAW文件在感光度测试中的应用。该手册是针对图像质量测试的专业参考资料，对从事摄影、图像处理、光学系统设计或质量控制的人员具有很高的价值。

Contrast sensitivity function

Noisy images

In images with long transitions (10-90% risetime r

1090

over 2 pixels) and high noise, increasing the noise can

increase the SQF, unless we take preventative steps. What we do is to recognize that most signal energy is at spatial

frequencies below 1/r

1090

. At frequencies over 1/r

1090

we do not allow MTF to increase: MTF( f

) = min( MTF(

n-1

), MTF( f

) ). This has little or no effect on the MTF due to the edge, but prevents noise spikes from unduly

increasing SQF.

Contrast sensitivity function CSF

The human eye's contrast sensitivity function (CSF) is limited by

the eye's optical system and cone density at high spatial (or

angular) frequencies and by signal processing in the retina

(neuronal interactions; lateral inhibition) at low spatial frequencies.

Various studies place the peak response at bright light levels

(typical of print viewing conditions) between 6 and 8 cycles per

degree. We have chosen a formula, described below, that peaks just

below 8 cycles/degree.

You may learn something about your own eye's CSF by viewing the

chart below at various distances and observing where the pattern

appears to vanish. Chart contrast is proportional to (y/h)

, for image

height h. To my eyes (which underwent Lasik refractive surgery in

1998 to correct for 10 diopter nearsightedness) it appears that the

peak in the curve on the right should be considerably broader. But

this isn't quite the case because of the eye's nonlinear response to

contrast. Although the chart below appears to decrease in contrast

linearly from top to bottom, the middle of the chart has 1/4 the

contrast as the top.

Log frequency-Contrast chart, created by Test Charts.

We have chosen a formula that is relatively simple, recent, and provides a good fit to data. The source is J. L.

Mannos, D. J. Sakrison, ``The Effects of a Visual Fidelity Criterion on the Encoding of Images'', IEEE Transactions

on Information Theory, pp. 525-535, Vol. 20, No 4, (1974), cited on this page of Kresimir Matkovic's 1998

PhD thesis.

Imatest Documentation

16 of 451

CSF( f ) = 2.6 (0.0192 + 0.114 f ) exp(-0.114 f )

1.1

The 2.6 multiplier drops out of SQF when the normalization constant K is applied. This equation can be simplified

somewhat.

CSF( f ) = (0.0192 + 0.114 f ) exp(-0.1254 f )

The preferred SQF equation, SQF = K ∫CSF( f ) MTF( f ) d(log f ) = K ∫(CSF( f ) MTF( f ) / f ) d f , blows up when

the dc-term (0.0192) is present. Fortunately, as the above plot shows, removing it makes very little difference. The

formula used to calculate SQF in the preferred equation is CSF( f ) = 0.114 f exp(-0.1254 f ).

More geek stuff (additional equations, not used, but can be selected)

Some additional equations are included in the SQF options for experimental study.

SQF = K ∫CSF( f ) MTF( f ) d f for all angular frequencies f

SQF = K log( ∫CSF( f ) MTF( f ) d f ) for all angular frequencies f

where

K = 100% / ∫ CSF( f ) d f or K = 100% / log(∫ CSF( f ) d f ) is the normalization constant.

log (often written ln) is the natural logarithm, i.e., log

These equations were developed to address concerns about d(log f ) = d f /f , which becomes very large at low spatial frequencies. Integrals

containing this term blow up if CSF( f ) contains a dc (constant) term. Fortunately, as noted above, removing the dc term from CSF makes

very little difference and works quite well, so we only keep these equations for experimental purposes.

Granger used d(log f ) because he noticed that a constant percentage change in MTF corresponded to a just noticeable difference, i.e., the eye

responds logarithmically. Equation 1 is linear. Equation 2 is logarithmic, but presents some scaling problems: What to do about log(1) =

0? Should we use log

(comparable to density measurements), log

(comparable to exposure value or f-stop measurements), or log

(in

accordance with Boulder, Colorado community standards, where only organic, natural logarithms are employed)? We've chosen not to pursue

these equations for now.

Links

Bob Atkins' excellent introduction to MTF and SQF is highly recommended.

Bror Hultgren, with 26 years of experience as an imaging scientist at the Polaroid Corporation, has developed a number of tools

that relate SQF to viewer preferences, using category scaling and subjective testing. Bror is available for consulting.

Popular Photography has been using SQF for testing lenses for years. Their test results are well worth exploring.

Imatest Documentation

17 of 451

(A)

Const.

sensor

noise

(B)

Const.

pixel

noise

(C)

noise

(D)

Canon

EOS-10D

ISO 1600

Noise in photographic images

Introduction

Noise is a random variation of image density, visible as grain in film and pixel level variations in digital images. It is a key image

quality factor; nearly as important as sharpness. Since it arises from basic physics— the photon nature of light and the thermal

energy of heat— it will always be there. The good news is noise can be extremely low— often impreceptably low— in digital

cameras, particularly DSLRs with large pixels (5 microns square or larger). But noise can get ugly in compact digital cameras with

tiny pixels, especially at high ISO speeds.

In most cases noise is perceived as a degradation in quality. But some Black & White photographers like its graphic effect: Many

favor 35mm 35mm Tri-X film. The pointillist painters, most notably George Seurat, created "noise" (specks of color) by hand; a

task that can be accomplished in seconds today with Photoshop plugins.

But by and large, the majority of photographers, especially color and large-format photographers, dislike noise with good reason.

Noise is measured by several Imatest modules. Stepchart produces the most detailed results, but noise is also measured in

Colorcheck, SFR, and Light Falloff.

Appearance

The appearance of noise is illustrated in the stepchart images on the right. Noise is usually

measured as an RMS (root mean square) voltage. The mathematics of noise is presented in a

green box at the bottom of this page.

The stepcharts in columns (A)-(C) are simulated. They are assumed to have a minimum

density of 0.05 and density steps of 0.1, identical to the Kodak Q-13 and Q-14. They have

been encoded with gamma = 1/2.2 for optimum viewing at gamma = 2.2 (the

Windows/Internet standard). Strong noise— more than you'd find in most digital cameras—

has been added to columns (A) and (B). Column (C) is noiseless.

The fourth column (D) contains an actual Q-13 stepchart image taken with the Canon

EOS-10D at ISO 1600: a very ISO high speed. Noise is visible, but admirably low for such a

high ISO speed (thanks, no doubt, to software noise reduction).

The noise in (A) is constant inside the sensor, i.e., before gamma encoding. When it is

encoded with gamma = 1/2.2, contrast, and hence noise, is boosted in dark areas and

reduced in light areas. The Kodak publication, CCD Image Sensor Noise Sources,

indicates that this is not a realistic case. Sensor noise tends to increase with brightness.

The noise in (B) is uniform in the image file, i.e., its value measured in pixels is constant.

This noise must therefore increase with brightness inside the sensor (prior to gamma

encoding), and hence is closer to real sensor behaviour than (A). Noise appears relatively

constant except for the darkest zones, where it's not clearly visible. Noise is often lower in

the lightest zones, where a tonal response "S" curve superimposed on the gamma curve (or

saturation) reduces contrast, hence noise.

For this reason the middle zones— where noise is most visible— are used to calculate the

average noise: a single number used to characterize overall noise performance. We omit

zones where the density of the original chart (hence display density in an unmanipulated

image) is greater than 1.5 or less than 0.1.

Noise measurements

Noise measurements should ideally

Imatest Documentation

18 of 451

correlate with perceived appearance,

(in many cases) be referenced to the original scene so the measurement is not affected by the tonal response (contrast) of the

camera or RAW converter,

be detailed enough to allow accurate assessment of sensor and camera performance, and

be simple enough to interpret without difficulty.

Since noise is only meaningful in relationship to a signal, a Signal-to Noise Ratio (SNR or S/N) is often calculated. SNR can

be defined in many ways, depending on how Signal S is defined. For example, S can be an individual patch pixel level or the

pixel difference corresponding to a specified scene density range (1.45 or 1.5 are often used for this purpose). It is important to

know precisely how SNR is defined whenever it is discussed. SNR can be expresed as a simple ratio (S/N) or in decibels

(dB), where SNR (dB) = 20 log

(S/N). Doubling S/N corresponds to increasing SNR (dB) by 6.02 dB.

Since these requirements can be somewhat contradictory, Imatest modules have several noise and SNR measurements, some

simple and some detailed. Available displays in Stepchart and Colorcheck include:

Noise a function of pixel level or exposure, expressed in

Pixels, normalized to the pixel difference corresponding to a scene density range of 1.5 for Stepchart (comparable

to the density range of 1.45 for the GretagMacbeth ColorChecker) or the White - Black zones in the ColorChecker

(row 3, patches 1 - 6; density range = 1.45). Without this normalization, noise is a function of RAW converter response

curve; it increases with converter contrast.

Pixels, normalized to the maximum pixel value: 255 for 8/24-bit files.

Pixels (maximum of 255 for 8/24-bit files)

f-stops (or EV or zones; a factor of two in exposure), i.e., referenced to the original scene. Noise measured in f-stops

corresponds closely to human vision. See f-stop noise, below.

Noise as a simple (average) number

A good single number for describing noise is the average noise of the Y (Luminance) channel, measured in pixels

(normalized scene density difference of 1.5), shown in the lower plot in the figures, below. The lightest and darkest

zones, representing chart densities greater than 1.5 or less than 0.1 are excluded from the average.

S/N or SNR (dB) as a function of pixel level or exposure, where S is the pixel level of the individual test chart patch.

S/N or SNR (dB) as a simple (average) number.

A good single number for defining SNR is the White - Black SNR, SNR_BW, where S is the difference in pixel levels

between the White and Black patches in the GretagMacbeth ColorChecker in Colorcheck (density range of 1.45), or

the pixel difference corresponding to a chart (scene) density difference of 1.5 in Stepchart. Noise N is taken from a

middle level patch with density around 0.7.

SNR

= 20 log

(( S

WHITE

– S

BLACK

) / N

mid

)

a noise spectrum plot, described below.

In addition, Light Falloff displays a spatial map of noise.

Characteristic Stepchart results are shown below.

Column (B): Uniform pixel noise

Imatest Documentation

19 of 451

剩余450页未读，继续阅读

devilthelover

粉丝: 128

Imatest图像质量检测手册

Imatest介绍说明文档

Imatest使用说明文档

Imatest 教程

imatest图像测试培训书 imatest说明书

Imatest官方使用说明文档.zip

Imatest详细使用教程以及图像质量测试原理方法说明

Imatest新手必读：图像测试基础技巧快速掌握手册（2023年最新指南）

图像分辨率测试：Imatest关键指标深度解析

Imatest动态范围测试：应用场景与必备知识

Imatest脚本高级技巧：编写高效可复用测试脚本（专业指南）

动态范围优化手册：高通MSM8996 ISP调优新技巧

imatest下载

Imatest使用说明.zip

Imatest的使用教程

Imatest3.9的最新说明文档

Imatest 详细教程

Imatest简单教程

开发界面语义化：声控 + 画图协同生成代码.doc

LABVIEW与三菱PLC通信：实现数据批量读写的高效库解决方案

SQL Server 2005 使用Transaction

LCD电子时钟报时 （团队项目） 代码完整

最新资源

LCD电子时钟报时（团队项目）代码完整