Description • Prerequisites • Installation • Exercises 0X • Segmentation • Exercises 1X

Exercises for the subject Image processing, analysis and classification of the EIT Digital Data Science Master at UPM (Technical University of Madrid).

Exercises have three parts:

• Exercises 0X: From exercise 2 to 9.
• Segmentation exercises.
• Exercises 1X: Exercise 11 and 13.

For running the exercises the prerequisites are:

• Java version 1.8.
• Maven, if not installed visit install Maven.
• opencv version 4.3.0, if not installed for java visit this guide (Only needed for exercise segmentation 2b, wheel teeth count)

For compiling the exercises, and generate the jar file containing all the exercises you have to follow this steps:

mvn package
chmod +x assignment/**/*.sh

If you want to execute Segmentation exercise 2B (Wheel teeth count), you have to set the variable OPENCV_PATH in the script assignment/utils.sh.

The project is divided in the following folders:

• assignment/: Is the folder containing all the responses for the assignments. Is where all the scripts of the exercises and pdfs are located.
• report/: Is where the source code (.tex) for the reports is located.
• src/: Folder containing the Java source code of the exercises.
• test-images/: Images for testing each exercise.

In this section is explained how to execute each exercise:

NOTE: For executing the exercise is supposed to be located in the parent directory of the project.

• Exercise 2a

assignment/exercises-0x/exercise-2/exercise_02a_thresh.sh exercise_02a_input_01.pgm value exercise_02a_output_01.pgm

• Exercise 2b

assignment/exercises-0x/exercise-2/exercise_02b_compare.sh exercise_02b_input_01.pgm exercise_02b_input_02.pgm exercise_02b_output.txt

• Exercise 2c

Check exercise_02c.pdf file.

• Exercise 2d inf

assignment/exercises-0x/exercise-2/exercise_02d_inf.sh exercise_02d_input_01.pgm exercise_0d_input_02.pgm exercise_02d_output_01.pgm

• Exercise 2d sup

assignment/exercises-0x/exercise-2/exercise_02d_sup.sh exercise_02d_input_01.pgm exercise_0d_input_02.pgm exercise_02d_output_01.pgm

• Exercise 3a

assignment/exercises-0x/exercise-3/exercise_03a_erosion.sh i exercise_03a_input_01.pgm exercise_03a_output_01.pgm

• Exercise 3b

assignment/exercises-0x/exercise-3/exercise_03b_dilation.sh i exercise_03b_input_01.pgm exercise_03b_output_01.pgm

• Exercise 4a

assignment/exercises-0x/exercise-4/exercise_04a_opening.sh i exercise_04a_input_01.pgm exercise_04a_output_01.pgm

• Exercise 4b

assignment/exercises-0x/exercise-4/exercise_04b_closing.sh i exercise_04b_input_01.pgm exercise_04b_output_01.pgm

• Exercise 5a

assignment/exercises-0x/exercise-5/exercise_05a_idempotence_opening.sh i exercise_05a_input_01.pgm

• Exercise 5b

assignment/exercises-0x/exercise-5/exercise_05b_idempotence_closing.sh i exercise_05b_input_01.pgm

• Exercise 6a

assignment/exercises-0x/exercise-6/exercise_06a_closing_opening.sh i exercise_06a_input_01.pgm exercise_06a_output_01.pgm

• Exercise 6b

assignment/exercises-0x/exercise-6/exercise_06b_opening_closing.sh i exercise_06b_input_01.pgm exercise_06b_output_01.pgm

• Exercise 7a

assignment/exercises-0x/exercise-7/exercise_07a_idempotence_clo_ope.sh i exercise_07a_input_01.pgm

• Exercise 7b

assignment/exercises-0x/exercise-7/exercise_07b_idempotence_ope_clo.sh i exercise_07b_input_01.pgm

• Exercise 8

assignment/exercises-0x/exercise-2/exercise_08_filter.sh exercise_08_input_01.pgm

• Exercise 9

Check exercise_09.pdf file.

assignment/segmentation/segmentation_1a.sh operation size input.png output.png

The arguments means:

• operation: the type of operation could be erosion or dilation.
• size: size of the structuring element used for operation.
• input.png: input path for the image used.
• output.pgm: output path for the image.

As example:

assignment/segmentation/segmentation_1a.sh erosion 1 test-images/segmentation/hitchcock.png out/hitchcock.pgm

assignment/segmentation/segmentation_1b.sh connectivity size input.png output.png

The arguments means:

• connectivity: the type of connectivity could be 4-connectivity or 8-connectivity.
• size: size of the structuring element used for operation.
• input.png: input path for the image used.
• output.pgm: output path for the image.

As example:

assignment/segmentation/segmentation_1b.sh 4 1 test-images/segmentation/particles.png out/particles.pgm

This program does not have arguments, the output is the number of counted wheel tooth:

assignment/segmentation/segmentation_2a.sh

Expected output:

Number of teeth > 102

Note: For this exercise the path of the opencv installation should be set in the assignment/utils.sh file.

This exercise does not expect any argument:

assignment/segmentation/segmentation_2b.sh

The used image is: test-images/segmentation/coffee_grains.jpg and it will store the generated markers in folder: out/markers.pgm

Implementation of the watershed algorithm could be found on location src/main/java/utils/segmentation/fast/FastWaterShed.java

The program could be executed with:

assignment/segmentation/segmentation_3a.sh input.pgm output.png

Where the arguments stand for:

• input.pgm: input image where regions are going to be extracted from.
• output.png: output image where regions are going to be stored in different colors.

An example of the execution with small image test-image/segmentation-test.pgm is already in the following script:

assignment/segmentation/segmentation_3a_example.sh

Also the pseudocode report could be found in

assignment/segmentation/segmentation_3a_pseudocode.pdf

• Exercise 11

assignment/exercises-1x/exercise_11a_flatzone.sh exercise_11a_input_01.txt exercise_11a_input_01.pgm exercise_11a_output_01.pgm

• Exercise 13a

assignment/exercises-1x/exercise_13a_minimum.sh exercise_13a_input_01.txt exercise_11a_input_01.pgm exercise_13a_output_01.txt

• Exercise 13b

assignment/exercises-1x/exercise_13b_maximum.sh exercise_13b_input_01.txt exercise_13b_input_01.pgm exercise_13b_output_01.txt

• Cristian M. Abrante Dorta - CristianAbrante