upload a function to calculate IoU

projects/Compute_IoU/Compute_IoU.py

@@ -0,0 +1,33 @@
+import numpy as np
+
+def Cal_IoU(GT_bbox, Pred_bbox):
+    '''
+    Args:
+        GT_bbox:  the bounding box of the ground truth
+        Pred_bbox: the bounding box of the predicted
+    Returns:
+        IoU: Intersection over Union
+    '''
+    #1. Calculate the area of the intersecting area
+    ixmin = max(GT_bbox[0], Pred_bbox[0])
+    iymin = max(GT_bbox[1], Pred_bbox[1])
+    ixmax = min(GT_bbox[2], Pred_bbox[2])
+    iymax = min(GT_bbox[3], Pred_bbox[3])
+    iw = np.maximum(ixmax - ixmin + 1., 0.) # the weight of the area
+    ih = np.maximum(iymax - iymin + 1., 0.) # the height of the area
+    area = iw * ih
+
+    #2. Calculate the area of all area
+    #S = S1 + S2 - area
+    S1 = (Pred_bbox[2] - GT_bbox[0] + 1) * (Pred_bbox[3] - GT_bbox[1] + 1)
+    S2 = (GT_bbox[2] - GT_bbox[0] + 1) * (GT_bbox[3] - GT_bbox[1] + 1)
+    S = S1 + S2 - area
+
+    #3. Calculate the IoU
+    iou = area / S
+    return iou
+
+if __name__ == "__main__":
+    pred_bbox = np.array([40, 40, 100, 100])
+    gt_bbox = np.array([70, 80, 110, 130])
+    print(Cal_IoU(pred_bbox, gt_bbox))

projects/Compute_IoU/README.md

@@ -0,0 +1,19 @@
+
+# Image-Watermark
+
+## Description
+
+The project will help you calculate Intersection over Union (IoU).
+
+## About this Project
+
+Intersection over Union (IoU) is used when calculating mAP. It is a number from 0 to 1 that specifies the amount of overlap between the predicted and ground truth bounding box.
+
+## Usage
+
+Use the Script [Compute_IoU.py](https://github.com/Python-World/python-mini-projects/blob/master/projects/Compute_IoU/Compute_IoU.py) . In the command line, Enter
+
+`python3 Compute_IoU.py `
+
+## Author
+[Mason](https://github.com/JohnMasoner)