Highlights
1) Transformations
Rotation, translation and skew are useful operations for matching, tracking, and data augmentation.
a) Write a function that takes as input an image I, rotates it by an angle θ1 and horizontally skews it by an angle, θ2. Write the matrix formulation for image rotation R(.) and skewing S(.). Define all the variables. Note that the origin of the coordinate system of the programming environment you use might be different from the one shown in the lectures.
b) Create an image that contains your name written in Arial, point 72, capital letters. Rotate clockwise the image you created by 30, 60, 120 and -50 degrees.
Skew the same image by 10, 40 and 60 degrees.
Complete the process so that all the pixels have a value. Discuss in the report the advantages and disadvantages of different approaches.
c) Analyse the results when you change the order of the two operators, that is R(S(I)) and S(R(I)). i) Rotate the image by θ1 = 20 clockwise and then skew the result by θ2 = 50.
ii) Skew the image by θ2 = 50 and then rotate the result by θ1 = 20 clockwise.
Are the results of (i) and (ii) the same? Why?
Texture Descriptors and Classification (Use Datasets A and B)
The Local binary pattern (LBP) operator describes the surroundings of a pixel by generating a bit-code from the binary derivatives of a pixel.
a) Write a function that divides an input (greyscale) image into equally sized non-overlapping windows and returns the feature descriptor for each window as distribution of LBP codes. For each pixel in the window, compare the pixel to each of its 8 neighbours. Convert the resulting bit-codes (base 2) to decimals (base 10 numbers) and compute their histogram over the window. Normalize the histogram. The histogram is now a feature descriptor representing the window. Show in the report the resulting images and discuss how you handled the image border.
b) Come up with a descriptor that represents the whole image as consisting of multiple windows. For example, you could combine several local descriptions into a global description by concatenation. Discuss in the report alternative approaches. Using the global descriptor you created, implement a classification process that separates the images in the dataset into two categories: face images and non-face images (You could use simple histogram similarities). Comment the results in the report. Is the global descriptor able to represent whole images of different types (e.g. faces vs. cars)? Identify problems (if any), discuss them in the report and suggest solutions.
c) Decrease the window size and perform classification again. Comment the results in the report. d) Increase the window size and perform classification again. Comment the results in the report. e) Discuss how LBP can be used or modified for the analysis of dynamic textures in a video.
Object segmentation and counting (Use Dataset C)
Moving objects captured by fixed cameras are the focus of several computer vision applications.
a) Write a function that performs pixel-by-pixel frame differencing using, as reference frame, the first frame of an image sequence. Apply a classification threshold and save the results to disk.
b) Repeat the exercise using the previous frame as reference frame (use frame It-1 as reference frame for frame It, for each t). Comment the results in the report.
c) Write a function that generates a reference frame (background) for the sequence using for example frame differencing and a weighted temporal averaging algorithm.
d) Write a function that counts the number of moving objects in each frame of a sequence. Generate a bar plot that visualizes the number of objects for each frame of the whole sequence. Discuss in the report the implemented solution, including advantages and disadvantages. How could the disadvantages be overcome?
This IT Assignment has been solved by our IT Experts at My Uni Paper. Our Assignment Writing Experts are efficient to provide a fresh solution to this question. We are serving more than 10000+ Students in Australia, UK & US by helping them to score HD in their academics. Our Experts are well trained to follow all marking rubrics & referencing style.
Be it a used or new solution, the quality of the work submitted by our assignment experts remains unhampered. You may continue to expect the same or even better quality with the used and new assignment solution files respectively. There’s one thing to be noticed that you could choose one between the two and acquire an HD either way. You could choose a new assignment solution file to get yourself an exclusive, plagiarism (with free Turnitin file), expert quality assignment or order an old solution file that was considered worthy of the highest distinction.
© Copyright 2026 My Uni Papers – Student Hustle Made Hassle Free. All rights reserved.