Chapter1
1. An image may be defined as a two-dimensional function, f(x,y), where x and y are spatial coordinates, and the amplitude of f at any pair of coordinates (x,y) is called the intensity or gray level of the image at that point.
2. Image processing includes image acquisition, image storage, image transmission and digital image processing.
3. Low level process involves primitive operations such as image preprocessing to reduce noise, contrast enhancement, and image sharpening.
4. Mid-level process involves tasks such as segmentation, description, and classification (recognition) of individual objects.
5. As for mid-level process, its inputs are images, but its outputs are attributes extracted from those images.
6. Digital image processing encompasses processes whose inputs and outputs are images and, in addition, encompasses processes that extract attributes from images, up to and including the recognition of individual objects.
7. Image restoration is based on mathematical or probabilistic models of image degradation.
8. Image compression is to reduce the storage required to save an image, or the bandwidth required to transmit it.
9. Morphological processing is to extract image components that are useful in the representation and description of shape.
10. Which of the following can highlight certain features of interest of an image? (A) Image enhancement (B) Image restoration (C) Image compression (D) Image Segmentation
Chapter2
1.在晚上光线低的情况下锥状细胞起主要作用。（x）
2. At low levels of illumination vision is carried out by the rods, whereas at high levels (showing better discrimination) vision is the function of cones.
1
3. The perceived brightness is a simple function of intensity. (x)
4. Which of the following shows the fact that the perceived brightness is a simple function of intensity? (A) Simultaneous Contrast (B) Mach bands (C) Optical illusions (D) Perspective Illusion
5. Digitizing the coordinate values is called sampling; Digitizing the amplitude values is called quantization.
6. Intuitively, spatial resolution is a measure of the smallest discernible detail in an image.
7. Quantitatively, spatial resolution can be stated in a number of ways, with line pairs per unit distance, and dots (pixels) per unit distance.
8. Intensity resolution similarly refers to the smallest discernible change in intensity level.
9. 下列两图的直方图相同。（√）
10. Which of the following is the histogram of the following image？
2
(A) (B)
(B) (D)
11. RGB颜色模型主要用于电脑监视器。（√） 12.下列哪种颜色模型主要用于彩色喷墨打印机？（） （A）RGB（B）CMY（C）HSB（D）YIQ
13. There are three commonly used image interpolation methods，which are Nearest Neighbor Interpolation, Bilinear Interpolation and Cubic Polynomial Interpolation. 14. The pixels having a city-block distance from (x,y) less than or equal to some value r form a diamond centered at (x,y).
15. The pixels with chessboard distance from (x,y) less than or equal to some value r form a square centered at (x,y).
16.图像的直方图基本上可以描述图像的概貌。下面所给的a、b、c、d四个直方图中，哪幅直方图对应的图像最清晰？（ ）
3
Chapter3
一、填空
1. Image negative is to enhance white or gray detail embedded in dark regions of an image, especially when the black areas are dominant in size.
2. Log transform is to expand the values of dark pixels in an image while compressing the higher-level values.
3. Contrast stretching is used to expand the range of intensity levels in an image so that it spans the full intensity range of the recording medium or display device.
4. Low-contrast images can result from poor illumination, lack of dynamic range in the imaging sensor, or even the wrong setting of a lens aperture during image acquisition. 5. Intensity-level slicing can enhance features such as masses of water in satellite imagery and enhancing flaws in X-ray images.
6. Image subtraction is used to enhance differences between images 7. Image averaging can decrease additional noises.
8. Edges and other sharp intensity transitions (such as noise) in an image contribute significantly to the high-frequency content of its Fourier transform.
9. A notch filter rejects (or passes) frequencies in a predefined neighborhood about the center of the frequency rectangle 二、选择
1. Which of the following can highlight a specific range of intensities? (A) Contrast stretching (B) Bit-plane slicing (C) Intensity-level slicing (D) Log transformations
2. 用局部平均法对图像进行平滑处理后，噪声方差为处理前的（）倍
(A) M (B) 1/M (C)
M (D) 1/M
3. 对下面窗口内的像素进行中值滤波的结果是（）
212200198206202201 208205207(A) 205 (B) 202 (C) 206 (D) 207
4
4.关于图像的一阶和二阶微分，下列说法错误的是（） （A）一阶微分处理通常会产生较宽的边缘； （B）二阶微分处理对背景有较强的响应；
（C）一阶微分处理一般对灰度阶梯有较强的响应； （D）二阶微分处理对灰度阶梯变化产生双响应。 5.下列哪个是Sobel算子？
6.下列哪一项不是拉普拉斯算子？
(A) （B） （C） （D）
（A） （B） （C） （D）
7. 下列哪个算子是二阶微分算子？（ ） （A）Laplace算子 （B）Prewitt算子 （C）Sobel算子 （D）Roberts算子 8.下列哪个模板是方向模板？（ ）
??1?1?1???1?1?1?? （B）?222? ?18?1（A）?????????1?1?1????1?1?1????1?12??11?1?? （D）?1?2?1? ?12?1（C）????????2?1?1???11?1??9. 下列方法中，哪一项不能用来进行图像锐化处理？（ ） （A）高频加强滤波 （B）反锐化掩模法
5
（C）微分法 （D）中值滤波法 10. 下列哪个算子对噪声最敏感？（ ） （A）Laplace算子 （B）Prewitt算子 （C）Sobel算子 （D）Roberts算子
11. 下列方法中，哪一项不能用来进行图像平滑处理？（ ） （A）局部平均法 （B）反锐化掩模法 （C）低通滤波法 （D）中值滤波法 12. 下列哪个模板不属于方向模板？（ ）
11??1?1?1??1? （B）??1?21? 1?2?1（A） ???????1??11???1?11???2?1?1??11?1?? （D）?1?2?1? ?12?1（C）?????????1?12???11?1??三、判断
1. 直方图均衡化实质上是减少图像的灰度级以换取对比度的加大。(√) 2.中值滤波在抑制图像随机脉冲噪声方面非常有效。（√）
3.用多帧平均对图像进行平滑处理后，信噪功率比减少M倍，噪声方差增加M倍。（?）
4.图像锐化目的是突出图像中的细节或者增强被模糊了的细节。（√）
5. 一般来说，二阶微分处理比一阶微分好一些，因为形成增强细节的能力好一些。（√） 四、简答
1. Under what circumstances will low-contrast images produce？ (1) Poor illumination;
(2) Lack of dynamic range in the imaging sensor;
(3) The wrong setting of a lens aperture during image acquisition. 2. 简述图像频域滤波处理的步骤。
（1）用(-1)x+y乘以输入图像进行中心变换。 （2）计算图像的DFT F(u,v)。 （3）用滤波器函数H(u,v)乘以F(u,v)。
6
（4）计算（3）中结果的逆DFT。 （5）得到（4）中结果的实部。 （6）用 (-1)x+y乘以（5）中结果。 3. 在频域对图像进行锐化都有哪些方法？
（1）Ideal, Butterworth, Gaussian highpass filters; （2）The Laplacian;
（3）Unsharp Masking（反锐化掩模）, Highboost Filtering （高提升滤波）and High-Frequency-Emphasis Filtering（高频加强滤波）; （4）Homomorphic Filtering（同态滤波） 4. 灰度变换图像增强有哪些方法？各有什么用途？ 五、应用
1. 设图像有64*64=4096个象素，有8个灰度级，灰度分布如表所示。进行直方图均衡化。
2. Give a single intensity transformation function for spreading the intensities of an image so the lowest intensity is C and the highest is L-1. 3. Explain why the discrete histogram equalization technique does not, in general, yield a flat histogram.
4. Two images, f(x,y) and g(x,y), have histograms f(x,y)+g(x,y).
5. Consider a 5?5 spatial mask that averages the 12 closest neighbors of a point (x,y), but excludes the point itself from the average.
(a) Find the equivalent filter, H(u,v), in the frequency domain. (b)Show that your result is a lowpass filter.
6. Each filter in the figure has a strong spike（尖峰） in its center. Explain the source of these spikes.
rkr0=0r1=1/7r2=2/7 r3=3/7 r4=4/7 r5=5/7 r6=6/7 r7=1 nkp(rk)7901023850656329245122810.190.250.210.160.080.060.030.02hf and hg. Give the conditions under which you can determine the histograms of
7
Chapter4
1.下列哪一项不属于物理退化模型？ （A）一般点的线性退化 （B）空间模糊退化 （C）旋转运动引起的退化 （D）随机噪声引起的退化
2. The principal sources of noise: during image acquisition (digitization) and/or transmission.
3. 白噪声的傅里叶谱是常数。（√）
4. Which of the following cannot be used to estimate the Degradation Function? (A) Observation (B) Experimentation (C) Modeling (D) Computing
5. For image restoration in the presence of noise only，which of the following filters could obtain the best results？ （ ）
（A）Mean Filters （B）Order-Statistics Filters （C）Adaptive Filters （D）Notch Filters
6. Which of the following filters could be used to reduce periodic noise？ （ ） （A）Bandreject Filters （B）Bandpass Filters （C）Notch Filters （D）Mean Filters
8
7. 对于逆滤波均值滤波器，Q为正时用来消除 胡椒 噪声，Q为负时用来消除 盐 噪声。
8. 下列哪种滤波器能有效消除胡椒噪声？ （A）算术均值 （B）几何均值 （C）最大值滤波器 （D）最小值滤波器 9. 下列哪种滤波器能有效消除盐噪声？ （A）算术均值 （B）几何均值 （C）最大值滤波器 （D）最小值滤波器 10．针对图像退化/复原模型，讨论图像复原方法。 11.简述逆滤波图像复原的原理。 12. 物理退化模型有哪几种？
Chapter5 Medical Image Compression
1. Data compression is the process of reducing the amount of data required to represent a given quantity of information.
2. Various amounts of data can be used to represent the same amount of information, representations that contains irrelevant or repeated information are said to contain redundant data.
3. There are three kind of data redundancy, which are coding redundancy, spatial and temporary redundancy and irrelevant information.
4. 下表是一个变长编码的例子，编码前数据的相对冗余是多少？
rk 0 1/7 2/7 3/7 4/7 5/7 6/7 1
Pr(rk) 0.19 0.25 0.21 0.16 0.08 0.06 0.03 0.02 等长编码 000 001 010 011 100 101 110 111 9
l1(rk) 3 3 3 3 3 3 3 3 变长编码 11 01 10 001 0001 00001 000001 000000 l2(rk) 2 2 2 3 4 5 6 6
(A) 1.1 (B) 0.091 (C) 0.11 (D) 0.91
5.关于图像的保真度准则，下列说法错误的是（ ）
（A）均方根误差Erms依赖于图像强度对比。例如，Erms＝100对于一个8bit图像（象素值0～255）来说看起来很糟糕，但对于一个10bit图像（象素值0～1023）其畸变却难以观察到。
（B）均方信噪比SNRms用分贝表示。SNRms越低，图像质量越好。
（C）SNRms对于比较一幅图像与它的压缩结果来说是很好的度量，但是在不同图像之间比较SNRms没有意义。
（D）图像的保真度准则包括主观保真度准则和客观保真度准则。 6. 关于JPEG和JPEG2000标准，下列说法错误的是（）
（A）JPEG算法包括初始输入、DCT、量化、编码等步骤，每一步都对图像的整体压缩有贡献。
（B）JPEG主要用于有损压缩。
（C）JPEG2000压缩效率较JPEG提高大约5%—30%
（D）JPEG中采用的变换是FFT，JPEG2000中采用的是DCT。 7. Huffman编码的理论依据是变长编码理论。
8. 算术编码也是对出现概率较大的符号采用短码，对概率较小的符号采用长码。但是，它的编码原理与Huffman编码大不相同。（）
9. 假设信源符号为{00，01，10，11}，它们出现的概率分别为{0.1，0.4，0.2，0.3}。根据该条件对输入序列10 00 11 00 10进行算术编码， 下列哪一项可以作为编码输出？（ ）
（A）0.5143 （B）0.514 （C）0.5146 （D）0.5152
10. Lempel-Ziv-Welch coding assigns fixed-length code words to variable length sequences of source symbols.
11. 在LZW编码中，字典的大小是一个重要的系统参量。字典太大，灰度级序列匹配会变得不太可能；太小，码字的尺寸反而会影响其压缩性能。（）
12. Images with repeating intensities along their rows/columns can often be compressed by representing runs of identical intensities as run-length pairs. 13. 对下图4x4像素进行行程编码。
5 5 10
5 3 4 4 2 4 1 2 4 3 2 4 3 2 14. 简述医学图像压缩的必要性与可行性。 15. How many parts does a source encoder model include？Please draw the model, and then write out each part’s usage.
16.一幅图像有8个灰度级，各灰度在图像中出现的概率如下表所示，请对这些灰度进行Huffman编码。
rk 0 1/7 2/7 3/7 4/7 5/7 6/7 1
Pr(rk) 0.19 0.25 0.21 0.16 0.08 0.06 0.03 0.02 11