Description

Task 1A – Detect objects and enlist their properties

Goal:
To learn Python and use it to:
• Identify shape of objects.
• Identify colour of objects.
• Identify position of objects in the image

Please find the Task-1A_main.py file in the nested folder named “2. Task_DescriptionTest Images”. Modify the sections of Task-1A_main.py marked for the same, to accomplish the following:

Given:
A set of five images named as test1.png, test2.png, test3.png, test4.png and test5.png in a folder named as “Test Images”. A sample set is shown in Figure 1. This folder is a sub-folder inside the “Task_Description” folder.

Figure 1: Sample Image Set containing five test images

Every image has a certain number of objects. Each Object is defined by three features, viz. Color, Shape and Position in the image. Color can be one of three primary colors: Red, Green and Blue. Shape can be one of any of the six shapes, viz. Triangle, Rhombus, Trapezium, Pentagon, Hexagon and Circle. Objects in a given image will vary in color, shape and position in the image and can appear multiple times in an image.

Problem Description:

For each image in the “Test Images” folder, add to the Python script file your methods or functions in order to find the color, shape and X and Y coordinates of centroid (or center) of each object in that image.

Your code should return a List of Lists; where each individual List contains – image name followed by List of Color-Shape-centroidX-centroidY in a comma separated values List.

Each List must have as many elements of Color-Shape-centroidX-centroidY as number of objects in that image.

You have to find out the color, Shape and X and Y coordinates of centroid of each object in the image and concatenate their values into one string in the format given below then store it in the List.

Considering the image “Sample1.png”, shown in Figure 2, the output would be as follows:

[“sample1.png”, [“Green-Rhombus-191-361”], [“Blue-Circle-524-361”], [“Red-Triangle-
831-392”]]

Figure 2: Sample1.png

Required Output:
1. The Python script file named Task-1A_main.py with the function “main” populated with your logic of solving the task.
2. A csv file with the Lists per image for each of the images in the given set of Test Images on each row i.e. five rows of data for a total of five test images in the “Test Images” folder. Each image will contain a maximum of 5 shapes. For our example, Sample1.png, the csv file would look like:

Sample1.png green-rhombus-191-361 red-triangle-831-392 blue-circle-524-361

3. An output image as shown in Figure 5:

Figure 3: Example Output Image with all properties enlisted, overlaid on the object for Sample1.png image

Name the output images with respect to its corresponding input image as – “test1output.png” for an input image named “test1.png” and so on.

To do:

1. Open Task-1A_main.py located in the folder named “Test Images”. It has two functions:

• main(): It returns the required Lists for the writecsv() function. It should also write the output image to file. This is the method where you write your logic to solve the task.

• writecsv(): This function expects four parameters as arguments-Color, Shape,
CentroidX and CentroidY and writes the elements of List of Lists one at a time into a
“result1A_<TeamID>.csv” file.
Do not edit this function. However, you can edit the global variable – “filename” with your TeamId in the Task-1A_main.py Python code file. Please use your eYRC team id while actually naming the file.

IMPORTANT: Do not change names of any of these functions.

Rules:

1. You need to write a generic program. Your code should be capable enough to detect any number of objects in an image and enlist their mentioned properties vis. Color, Shape and Position. In addition – your code will be tested on several undisclosed images when you submit your code.

2. Use basic knowledge of geometry to differentiate between the shapes of objects.

3. Objects are not rotated or differently oriented with respect to each other.

4. In case object is not present in the image, return nothing or an empty List.

5. In case objects are found in the image, return the List of [Color-Shape-centroidXcentroidY] in addition with image name; where there is one List appended per object in the image and one List of Lists per image.

Happy Learning

All The Best!!!