OpenCV

Question 1

What is Template 2D point class?

Answer 1

Point_

Question 2

What is Template 3D point class?

Answer 2

Point3_

Question 3

What is Template size (width, height) class?

Answer 3

Size_

Question 4

What is Template short vector class?

Answer 4

Vec

Question 5

What is 4-element vector class?

Answer 5

Scalar

Question 6

What is Rectangle class?

Answer 6

Rect

Question 7

What is 2D dense array (used as both a matrix or an image) class?

Answer 7

Mat

Question 8

What is Multi-dimensional dense array class?

Answer 8

MatND

Question 9

What is Multi-dimensional sparse array class?

Answer 9

SparseMat

Question 10

What is Template smart pointer class?

Answer 10

Ptr

Question 11

Create a matrix with width=320, height=240

Answer 11

Mat image(240, 320, CV_8UC3);

Question 12

(Re)allocate a per-declared matrix

Answer 12

image.create(480, 640, CV_8UC3);

Question 13

Create a matrix initialized with a constant

Answer 13

// CV_32F: R, G, B values are floating numbers between 0 and 1.
Mat A33(3, 3, CV_32F, Scalar(5)); // Scalar is a 4-element vector.
Mat B33(3, 3, CV_32F);
B33 = Scalar(5);
Mat C33 = Mat::ones(3, 3, CV_32F)*5.;
Mat D33 = Mat::zeros(3, 3, CV_32F) + 5.;

Question 14

Create a matrix initialized with specified values

Answer 14

double a = CV_PI/3;
Mat A22 = (Mat_(2,2) **<< ** cos(a), -sin(a), sin(a), cos(a));

float B22data[] = {cos(a), -sin(a), sin(a), cos(a)};
Mat B22 = Mat(2, 2, CV_32F, B22data).clone();

Question 15

Initialize a random matrix

Answer 15

// uniform dist from 0 to 256
**randu**(image, **Scalar(0)**, **Scalar(256)**);

// Gaussian dist with mean=128, std=10
**randn**(image, **Scalar(128)**, **Scalar(10)**);

Question 16

Access matrix elements: Different ways!

Answer 16

// split the image into separate color planes
vector<Mat> planes;
split(img_yuv, planes);

// method 1. process Y plane using an iterator
 MatIterator**\_**\<uchar\> it = planes[0].begin\<uchar\>(),
                     it\_end = planes[0].end\<uchar\>();
 for(; it != it\_end; ++it)
 {
     double v = **\*it**\*1.7 + rand()
     **\*it** = **saturate\_cast\<uchar\>**(v\*v/255.);
 }

// method 2. process the first chroma plane using pre-stored row pointer.
 // method 3. process the second chroma plane using
 //           individual element access operations
 for( int y = 0; y \< img\_yuv.rows; y++ )
 {
     uchar\* Uptr = planes[1].ptr\<uchar\>(y);
     for( int x = 0; x \< img\_yuv.cols; x++ )
     {
         Uptr[x] = saturate\_cast\<uchar\>((Uptr[x]-128)/2 + 128);
         uchar& Vxy = planes[2].at\<uchar\>(y, x);
         Vxy = saturate\_cast\<uchar\>((Vxy-128)/2 + 128);
     }
 }

merge(planes, img_yuv);
…

Question 17

Copy matrix to another one

Answer 17

src.copyTo(dst); // src is a Mat object

Question 18

Make deep copy of a matrix

Answer 18

m.clone() // m is a Mat object

Question 19

Scale and convert to another datatype

Answer 19

src.convertTo(dst, type, scale, shift); // src is a Mat object

Question 20

Change matrix dimensions and/or number of channels without copying data

Answer 20

m.reshape(nch, nrows); // m is a Mat object

Question 21

Take a matrix ‘m’ row/column span

Answer 21

m. rowRange(Range(i1, i2));
m. colRange(Range(j1, j2));

Question 22

Take a matrix ‘m’ diagonal

Answer 22

m.diag(i)

Question 23

Take a submatrix of ‘m’

Answer 23

m(Range(i1, i2), Range(j1, j2));
m(roi);

Question 24

Make a bigger matrix from a smaller one ‘m’

Answer 24

m.repeat(ny, nx)

Question 25

Ex: Smooth image ROI in-place

Answer 25

* *Mat** *imgroi* = image(Rect(10, 20, 100, 100)); * *GaussianBlur**(*imgroi*, *imgroi*, Size(5,5), 1.2, 1.2);

Question 26

Simple Matrix Operations

Answer 26

add(), subtract(), multiply(), divide(), absdiff(), bitwise\_and(), bitwise\_or(), bitwise\_xor(), max(), min(), compare()

Question 27

Ex: Alpha compositing functions

Answer 27

void alphaCompose(**const** **Mat&** rgba1, **const** **Mat&** rgba2, **Mat&** rgba\_dest) { Mat a1(rgba1.size(), rgba1.type()), ra1; Mat a2(rgba2.size(), rgba2.type()); int mixch[]={3, 0, 3, 1, 3, 2, 3, 3}; mixChannels(&rgba1, 1, &a1, 1, mixch, 4); mixChannels(&rgba2, 1, &a2, 1, mixch, 4); subtract(Scalar::all(255), a1, ra1); bitwise\_or(a1, Scalar(0,0,0,255), a1); bitwise\_or(a2, Scalar(0,0,0,255), a2); multiply(a2, ra1, a2, 1./255); multiply(a1, rgba1, a1, 1./255); multiply(a2, rgba2, a2, 1./255); add(a1, a2, rgba\_dest); }

Question 28

Various statistics of matrix elements

Answer 28

sum(), mean(), meanStdDev(), norm(), countNonZero(), minMaxLoc()

Question 29

Classical math functions

Answer 29

exp(), log(), pow(), sqrt(), cartToPolar(), polarToCart()

Question 30

Image Processing: Filtering

Answer 30

filter2D(); // Non-separable linear filter sepFilter2D(); // Separable linear filter boxFilter(); GaussianBlur(); medianblur(); bilateralFilter(); Sobel(); Scharr(); Laplacian(); erode(); dilate();

Question 31

Example. Filter image in-place with a 3x3 high-pass kernel (preserve negative responses by shifting the result by 128):

Answer 31

filter2D(image, image, image.depth(), (Mat \(3,3)\<\<-1, -1, -1, -1, 9, -1, -1, -1, -1), Point(1,1), 128);

Question 32

Geometrical Transformations

Answer 32

resize(); // Resize image getRectSubPix(); // Extract an image patch warpAffine(); // Warp image affinely warpPerspective(); // Warp image perspectively remap(); // Generic image warping convertMaps(); // Optimize maps for a faster remap() execution

Question 33

Ex: Decimate image by factor of square root of 2

Answer 33

Mat dst; resize(src, dst, Size(), 1./sqrt(2), 1./sqrt(2));

Question 34

Various Image Transformations

Answer 34

cvtColor(); // Convert image from one color space to another threshold(); // Convert grayscale image to binary image adaptivethreshold(); // Using a fixed or a variable threshold floodFill(); // Find a connected component using region growing algorithm integral(); // Compute integral image distanceTransform(); // build distance map or discrete Voronoi diagram for a binary image\< watershed(); // marker-based image segmentation algorithms grabCut();

Question 35

Histograms

Answer 35

calcHist(); calcBackProject(); equalizeHist(); compareHist(); // Compare two histograms

Question 36

Ex: Histogram

Answer 36

**Mat** hsv, H; **MatND** tempH; cvtColor(image, hsv, CV BGR2HSV); int planes[]={0, 1}, hsize[] = {32, 32}; **calcHist**(&hsv, 1, planes, Mat(), tempH, 2, hsize, 0); H = tempH;

Question 37

Data I/O

Answer 37

XML/YAML storages are collections (possibly nested) of scalar values, structures and heterogeneous lists.

Question 38

Writing and reading raster images (The functions can read/write images in the following formats: BMP (.bmp), JPEG (.jpg, .jpeg), TIFF (.tif, .ti), PNG (.png), PBM/PGM/PPM (.p?m), Sun Raster (.sr), JPEG 2000 (.jp2). Every format supports 8-bit, 1- or 3-channel images. Some formats (PNG, JPEG 2000) support 16 bits per channel.)

Answer 38

* *imwrite**("myimage.jpg", image); * *Mat** image\_color\_copy = **imread**("myimage.jpg", 1); * *Mat** image\_grayscale\_copy = **imread**("myimage.jpg", 0);

Question 39

Reading video from a file or from a camera

Answer 39

**VideoCapture** cap; if(argc \> 1) cap.**open**(string(argv[1])); else cap.**open**(0); Mat frame; **namedWindow**("video", 1); for(;;) { **cap \>\> frame**; if(!frame.data) break; **imshow("video", frame);** if(waitKey(30) \>= 0) break; }

Question 40

Suppose 'frame' is a pointer to **IplImage** struct with **uchar** pixel type. How do you access the value of the pixel (x, y)?

Answer 40

pixel\_value\_B = (**(uchar \*)** (frame-\>imageData+y\*frame-\>widthStep))[x\*frame-\>nChannels+0]; pixel\_value\_G = (**(uchar \*)** (frame-\>imageData+y\*frame-\>widthStep))[x\*frame-\>nChannels+1]; pixel\_value\_R = (**(uchar \*)** (frame-\>imageData+y\*frame-\>widthStep))[x\*frame-\>nChannels+2];

Question 41

Conversion between (**IplImag**e or **CvMat**) and **cv::Mat**

Answer 41

\* From OpenCV 2.\*, **cv::Mat** replaces the **CvMat** and **IplImage**, but it's easy to convert between the old and the new data structures. For more on this, please see [this](http://opencv.willowgarage.com/documentation/cpp/introduction.html). ``` // Assuming somewhere **IplImage \*iplimg;** exists // and has been allocated and **cv::Mat Mimg** has been defined **Mat** imgMat(iplimg); //Construct an Mat image "img" out of an IplImage Mimg = iplimg; //Or just set the header of pre existing cv::Mat Ming // to iplimg's data (no copying is done) ``` //Convert to IplImage or CvMat, no data copying * *IplImage** ipl\_img = img; * *CvMat** cvmat = img; // convert cv::Mat -\> CvMat

Question 42

List the modules in OpenCV.

Answer 42

* **core** - a compact module defining basic data structures, including the dense multi-dimensional array Mat and basic functions used by all other modules. * **imgproc** - an image processing module that includes linear and non-linear image filtering, geometrical image transformations (resize, affine and perspective warping, generic table-based remapping), color space conversion, histograms, and so on. * **video** - a video analysis module that includes motion estimation, background subtraction, and object tracking algorithms. * **calib3d** - basic multiple-view geometry algorithms, single and stereo camera calibration, object pose estimation, stereo correspondence algorithms, and elements of 3D reconstruction. * **features2d** - salient feature detectors, descriptors, and descriptor matchers. * **objdetec**t - detection of objects and instances of the predefined classes (for example, faces, eyes, mugs, people, cars, and so on). * **highgui** - an easy-to-use interface to video capturing, image and video codecs, as well as simple UI capabilities. * **gpu** - GPU-accelerated algorithms from different OpenCV modules. * ... some other helper modules, such as **FLANN** and **Google test wrappers**, **Python bindings**, and others.