# Recent questions tagged digital-image-processing

1
I am a dropper and have worked an year on machine learning in the industry. I want to do learn more about ML/AI and image processing. I am getting a rank around 200(from GO rank predictor). I am open to 2 or 3 year courses and any college is fine as long as I get to work in my field of interest and the resources are good there. What are the best options for me?
2
64. A 4*4 DFT matrix is given by 1\2 1 1 1 1 1 x -1 y 1 -1 1 -1 1 -j -1 j (j2= -1) Where values of x and y are ---------,------------- respectively. (1)1,-1 (2)-1,1 (3)-j,j (4)j,-j
1 vote
3
The three aspects of Quantization, programmers generally concerned with are Coding error, Sampling rate and Amplification Sampling rate, Coding error and Conditioning Sampling rate, Aperture time and Coding error Aperture time, Coding error and Strobing
1 vote
4
If the histogram of an image is clustered towards origin on X-axis of a histogram plot then it indicates that the image is ______. Dark Good contrast Bright Very low contrast
1 vote
5
Which of the following is not used in standard JPEG image compression ? Huffman coding Runlength encoding Zig-zag scan K-L Transform
6
Consider a discrete memoryless channel and assume that H(x) is the amount of information per symbol at the input of the channel; H(y) is the amount of information per symbol at the output of the channel. H(x $\mid$ y) is the amount of uncertainty remaining on x knowing y; and I(x;y)is the information transmission. ... -H(y $\mid$ x)]; p(x) max [H(x)-H(x $\mid$ y)]; p(x) max H(x $\mid$y); p(x)
1 vote
7
Blind image disconvolution is Combination of blur identification and image restoration Combination of segmentation and classification Combination of blur and non-blur image None of the above
8
Consider the conditional entropy and mutual information for the binary symmetric channel. The input source has alphabet $X=\{0,1\}$ and associated probabilities ${\dfrac{1}{2}, \dfrac{1}{2}}$. The channel matrix is $\begin{pmatrix} 1-p & p \\ p & 1-p \end{pmatrix}$ wgere p is the transition probability. Then ... is given by: $1$ $-p \log(p)-(1-p) \log(1-p)$ $1+p \log(p)+(1-p) \log(1-p)$ $0$
9
A Butterworth lowpass filter of order $n$, with a cutoff frequency at distance $D_{0}$ from the origin, has the transfer function $H(u, v)$ given by $\frac{1}{1+\left[\frac{D(u, v)}{D_{0}}\right]^{2n}}$ $\frac{1}{1+\left[\frac{D(u, v)}{D_{0}}\right]^{n}}$ $\frac{1}{1+\left[\frac{D_{0}}{D(u, v)}\right]^{2n}}$ $\frac{1}{1+\left[\frac{D_{0}}{D(u, v)}\right]^{n}}$
1 vote
10
Given a simple image of size $10 \times 10$ whose histogram models the symbol probabilities and is given by $P_{1}$ $P_{2}$ $P_{3}$ $P_{4}$ a b c d The first order estimate of image entropy is maximum when $a = 0, b = 0, c = 0, d = 1$ $a=\frac{1}{2}, b=\frac{1}{2}, c=0, d=0$ $a=\frac{1}{3}, b=\frac{1}{3}, c=\frac{1}{3}, d=0$ $a=\frac{1}{4}, b=\frac{1}{4}, c=\frac{1}{4}, d=\frac{1}{4}$
11
Given two spatial masks $S_{1}= \begin{bmatrix} 0&1&0 \\ 1&-4&0 \\ 0&1&0 \end{bmatrix}$ $S_{2} = \begin{bmatrix} 1&1&1 \\ 1&-8&1 \\ 1&1&1 \end{bmatrix}$ The Laplacian of an image at all points $(x, y)$ can be implemented by convolving the image with spatial mask. Which of the following can be used as the spatial mask ? only $S_{1}$ only $S_{2}$ Both $S_{1}$ and $S_{2}$ None of these
12
The redundancy in images stems from pixel decolleration pixel colleration pixel quantization image size
13
Consider the formula in image processing $R_D = 1- \frac{1}{C_R}$ where $C_R = \frac{n_1}{n_2}. \quad C_R$ is called as compression ratio $n_1$ and $n_2$ denotes the number of information carrying units in two datasets that represent the same information. In this situation $R_D$ is called as relative ____ of the first data set. Data Compression Data Redundancy Data Relation Data Representation
1 vote
14
MPEG involves both spatial compression and temporal compression. The spatial compression is similar to JPEG and temporal compression removes _____ frames. Temporal Voice Spatial Redundant
15
If f(x, y) is a digital image, then x, y and amplitude values of f are Finite Infinite Neither finite nor infinite None of the above
16
The number of distinct binary images which can be generated from a given binary image of right M $\times$ N are M+N M $\times$ N 2$^{M+N}$ 2$^{MN}$
17
Images tend to be very large collection of data. The size of memory required for a 1024 by 1024 image in which the color of each pixel is represented by a n-bit number, (in an 8-bit machines) is n $\times$ 8 MB n/8 MB (1024 $\times$ 1024)/8 MB 1024 MB
1 vote
18
An image is 1024$*$800 pixels with 3 bytes/pixel. Assume the image is uncompressed. How long does it make to transmit it over a 10 Mbps Ethernet? 196.6 seconds 19.66 seconds 1.966 seconds 0.1966 seconds
19
The transform which possess the highest ‘energy compaction’ property is Slant transform Cosine transform Fourier transform Karhunen-Loeve transform
20
21
Which of the following is not an image type used in MPEG? A frame B frame D frame P frame
22
Which of the following compression algorithms is used to generate a .png file? LZ78 Deflate LZW Huffman
23
What does a pixel mask mean? string containing only 1's string containing only 0's string containing two 0's string containing 1's and 0&rsquo;s
24
25
14. If the pixels of an image are shuffled then the parameter that may change is (A) Histogram (B) Mean (C) Entropy (D) Covariance
If the frame buffer has 8 bits per pixel and 8 bits are allocated for each of the $R, G, B$ components, what would be the size of the lookup table? $24$ bytes $1024$ bytes $768$ bytes $256$ bytes
If the Fourier transform of the function f(x,y) is F(m,n), then the Fourier transform of the function f(2x,2y) is $\frac{1}{4} F (\frac{m}{2}, \frac{n}{2})$ $\frac{1}{4} F (2m, 2n)$ $\frac{1}{4} F (m,n)$ $\frac{1}{4} F (\frac{m}{4}, \frac{n}{4})$