一.填空题
1.The ensemble averages will be identical (with) the statistical averages and may be represented by the same symbols.
2.Suppose, for example, that the statistical characteristics of the sample functions (in) the ensemble were changing with time.
3.There is a negative statement associated (with) Shannon's theorem.
4.It turns out (that) the results obtained for a gaussian channel often provide a lower bound (on) the performance of a system operating (over) a nongaussian channel.
5.The characteristics of the pulse group are related (to) the message sample through operations called quantization and coding.
6.The term symbol coder distinguishes this coding operation (from) the overall source encoding process.In most cases,a variable_length code is used to represent the mapped and quantized date set.It assigns the shortest codewords (to) the most frequently (occurring) output values and thus reduces coding redundancy.
7.Quantization and sampling produce the same result (as) sampling and quantizing.
8.Error control coding waveforms,(on) the other hand, have bandwidth expansion factors that grow only linearly (with) the code block length.
9.The method involves first sampling the information signal,quantizing the sample by rounding (off) to the closest of a number of discrete levels, and finally generating a prescribed number of pulses according to a code related to the nearest discrete level.
10.Speech coding is a fundamental technology that has existed (for) more than 60 years,beginning (in) the1930s with Dudley’s original vocoder.
11Channel coding protects digital date (from) errors by selectively introducing redundancies in the transmitted data.
12.Vision is the most advanced of our senses, so it is not surprising that images play the signal most important role (in) human perception.
13.Sometimes a distinction is made by defining image processing (as).
14.A low-level process is characterized (by) the fact that both its inputs and outputs are images.
二.英译汉
1.random process. 随机过程 2.random variable 随机变量
3.channel capacity. 信道容量 4.error probability 误码率
5.positive pulse. 正脉冲 6.pulse code modulation脉冲编码调制
7.quantum step size. 量化步长 8.linear function线性函数
9.pulse train. 脉冲序列 11.error correction code. 纠错码 13.carrier frequency. 载波频率 15.Speech coding. 语音编码 强
17.computer display 计算机视觉
三.汉译英
1.样本函数 sample function 3 高斯信道 gaussian channel 5 下界，下限 lower bound 检错码
12.bandpass signal 带通信号
14.background noise背景噪声
16.contrast enhancement对比度增2集平均 ensemble average
4 概率密度 probability density
6 上限 upper limit
10.error detection code
7脉幅调制 PAM 8 脉宽调制. PDM
9 脉位调制 PPM 10 信道编码channel coding
11 加性高斯白噪声 AWGN 12 编码调制 state diagram
13 已调信号 vector array 14 铜线. copper wire
15 语音识别 speech recognition 16 电磁波频谱 electromagnetic spectrum
17 无线电波 radio wave 18 图像理解 image understanding
19 人工智能. artificial intelligence
四.翻译句子
1.在整篇文章中，我们假定要研究的随机过程均为各太历经过程。
Throughout this text we shall assume that the random processes with which we shall have occasion to deal are ergodic.
2.this negative theorem states that if the information rate R exceeds a specified value C,the error probability will increase toward unity as M increases, and that also, generally, in this case where R＞C ,increasing the complexity of the coding results in an increase in the probability of error.
这个逆定理表明，如果信息速率R超过规定值C，那么误码率将随着M的增加不断向一增加，通常，当R＞C时，编码复杂度的增加还会导致误码率的增加。
3.量化实际上是选择一组离散电平(如例题中选为8，间隔为1伏特)并将原采样值舍入到最近的离散或量化电平。
Quantization amounts to selecting a set of discrete voltage levels (say eight in the example,1 volt apart) and rounding off the exact samples to the nearest discrete or quantum level.
4.Clearly,information is lost in the process of rounding off exact sample values to the nearest quantum level.
很明显在将原采样值舍入为最近的离散或量化电平的过程中丢失了信息。
5.冗余比特的引入会增加链路中使用的原始数据速率，因而增加了固定信源数据速率所需的带宽。
The introduction of redundant bits increases the raw data rate used in the link, hence increases the bandwidth requirement for a fixed source data rate.
6.信道编码器对数字消息(或信源)数据编码，将信源信息编成适于信道传输的码序列。
A channel coder operates on digital message (or source) date by encoding the source information into a code sequence for transmission through the channel.
7.连续波调制和脉冲调制都可归类为模拟调制，其调制参数随着信息信号的线性函数连续变化。
Both cw and pulse modulation may be classed as analog modulation which involves varying the modulated parameter continuously as a linear function of the information signal.
8.As a case in point, efficient line-of-sight radio propagation requires antennas whose physical dimensions are at least 1/10of thesignal’s wavelength.
恰如此例，无线电的有效视距传播要求天线的实际尺寸至少为信号波长的十分之一。
9.Although this coding rate is more than acceptable for telecommunications,it limits the quality of the speech (to what we call telephone-quality speech) and,therefore,affects the performance of not just speech coders,but also speech recognition systems.
虽然这个编码速率超过了电信的要求，但它限制了语音质量(我们称之为电话语音质量)，因此不仅影响了语音编码器的性能，还影响了语音识别系统的性能。
10.The resulting coded wideband speech not only sounds better than telephone bandwidth speech, but is also more intelligible for humans and works well with modern speech recognition systems.
得到的宽带编码语音不仅听起来比电话带宽的语音好，而且更清晰，并适用于现代语
音识别系统。
11.从图像处理到计算机视觉并没有一个明确的分界。
There are no clear-cut boundaries in the continuum from image processing at one end to computer Vision at the other.
12.Based on the preceding comments, we see that a logical place of overlap between image processing and image analysis is the area of recognition of individual regions or objects in an image.
基于前面的说明，我们明白了图像处理和图像分析的逻辑重叠区是图像中的单个区域识别或目标识别。
因篇幅问题不能全部显示，请点此查看更多更全内容