The task is given in the PDF

• The code for the task is in two files, main.py and functions.py.
• All other code is in the directory randomcode, which is either stuff I wrote when learning OpenCV, or code I used to debug and test (showimage.py and detectShapes.py)

1. CASUALTY_ASSIGNMENT_DEETS: dict[str, dict[str, list[tuple[int, int]]]]

   • Example content:

   {
       "imageName1.png": {
           "blue": [(agegrp1, severity1), ...],
           "pink": [(agegrp2, severity2), ...],
           "gray": [(agegrp3, severity3), ...],
       }, 
       "imageName2.png": {
           "blue": [(agegrp4, severity4), ...],
           "pink": [(agegrp5, severity5), ...],
           "gray": [(agegrp6, severity6), ...],
       }, 
       ...
   }

2. CAMP_PRIORITY_SCORES_TOTAL: dict[str, dict[str, int]]

   • Example content:

   {
       "imageName1.png": {
           "blue": 12,
           "pink": 7,
           "gray": 10,
       },
       "imageName2.png": {
           "blue": 10,
           "pink": 15,
           "gray": 9,
       },
       ...
   }

3. CAMP_PRIORITY_SCORES_AVG: dict[str, float]

   • Example content:

   {
       "1.png": 4.833333333333333,
       "2.png": 4.25,
       ...
   }

4. CASUALTIES: list[tuple[tuple[int, int], int, int]]

   • Specific to each imag
   • Given to get_humans_and_camps to populate, stores the coordinates of the centroids of the casualty markers, the age group, and the severity of the casualty. Example:

   [((102, 485), 3, 3),
    ((460, 422), 3, 2),
    ((266, 416), 1, 3),
    ((262, 278), 2, 2),
    ((337, 132), 3, 1),
    ((192, 48), 2, 2)]

5. CAMPS: dict[str, tuple[int, int]]

   • Specific to each image
   • Given to get_humans_and_camps to populate, stores the coordinates of the centroids of the camps. Example:

   {'blue': (57, 338), 'gray': (367, 215), 'pink': (81, 94)}

6. PINK_POINTS, GRAY_POINTS, BLUE_POINTS: list[float]

   • Returned from get_priority_for_all_points
   • Contains relative points wrt distance from camp for a casualty.

7. FINAL_CAMP_ASSIGNMENT: dict[str, list[tuple[tuple[int, int], int, int, float]]]

   • Stores coordinates of the casualty marker, age group, severity, relative points.
   • The dictionary's keys will be pink, gray and blue, and the value will be a list of the details of all the casualty markers, with their relative points.
   • Note that the info for the points will be stored in each list (FINAL_CAMP_ASSIGNMENT['pink'], FINAL_CAMP_ASSIGNMENT['blue'], FINAL_CAMP_ASSIGNMENT['gray']) at the same index as in CASUALTIES

The analysis of image is done using the functions in functions.py.

1. ground_overlay

   def ground_overlay(img: np.ndarray) -> None:
   ...

   • Detects the ground portion in the parameter image and applies a yellow cover over it. Example:

   Without overlay	With overlay

2. get_humans_and_camps

   def get_humans_and_camps(
       img: np.ndarray,
       camps: dict[str, tuple[int, int]],
       casualties: list[tuple[tuple[int, int], int, int]],
   ) -> None:
       ...

   • Detects the contours (triangles, squares, stars and circles) and puts them in their respective variable.

3. get_priority_for_all_points

   def get_priority_for_all_points(
       camps: dict[str, tuple[int, int]],
       casualties: list[tuple[tuple[int, int], int, int]],
   ) -> list[list[float]]:

   • Returns a list of 3 elements (1 for each camp) which themselves are lists of floating point values (relative priority scores for each casualty marker wrt the respective camp).

   • If the casualty marker is at $(x, y)$ and the respective camp is at $(x_1, y_1)$, and the casualty marker's age group is $a$ and severity is $s$ then the relative priority points is:

     $$\text{rel score} = \dfrac{a\cdot s}{\sqrt[4]{(x_1 - x) ^ 2 + (y_1 - y)^2}}$$

     $$\text{ie, rel score} = \dfrac{\text{age grp}\cdot\text{severity}}{\sqrt{\text{distance between the casualty and the camp}}}$$

Original	Output