Update data_aug.py

data_aug/data_aug.py

@@ -89,6 +89,107 @@ def __call__(self, img, bboxes):
 
         return img, bboxes
 
+class VerticalFlip(object):
+
+    """vertically flips the Image
+
+
+    Returns
+    -------
+
+    numpy.ndaaray
+        Flipped image in the numpy format of shape `HxWxC`
+
+    numpy.ndarray
+        Tranformed bounding box co-ordinates of the format `n x 4` where n is
+        number of bounding boxes and 4 represents `x1,y1,x2,y2` of the box
+
+    """
+
+    def __init__(self):
+        pass
+
+    def __call__(self, img, bboxes):
+        img_center = np.array(img.shape[:2])[::-1]/2
+        img_center = np.hstack((img_center, img_center))
+        img = img[::-1, :, :]
+        bboxes[:, [1, 3]] += 2*(img_center[[1, 3]] - bboxes[:, [1, 3]])
+
+        box_h = abs(bboxes[:, 1] - bboxes[:, 3])
+
+        bboxes[:, 1] -= box_h
+        bboxes[:, 3] += box_h
+
+        return img, bboxes
+
+
+class HorizontalFlip(object):
+
+    """horizontally flips the Image 
+
+    Returns
+    -------
+
+    numpy.ndaaray
+        Flipped image in the numpy format of shape `HxWxC`
+
+    numpy.ndarray
+        Tranformed bounding box co-ordinates of the format `n x 4` where n is
+        number of bounding boxes and 4 represents `x1,y1,x2,y2` of the box
+
+    """
+
+    def __init__(self):
+        pass
+
+    def __call__(self, img, bboxes):
+        img_center = np.array(img.shape[:2])[::-1]/2
+        img_center = np.hstack((img_center, img_center))
+
+        img = img[:, ::-1, :]
+        bboxes[:, [0, 2]] += 2*(img_center[[0, 2]] - bboxes[:, [0, 2]])
+
+        box_w = abs(bboxes[:, 0] - bboxes[:, 2])
+
+        bboxes[:, 0] -= box_w
+        bboxes[:, 2] += box_w
+
+        return img, bboxes
+
+class RandomVerticalFlip(object):
+
+    """Randomly vertically flips the Image with the probability *p*
+
+
+    Returns
+    -------
+
+    numpy.ndaaray
+        Flipped image in the numpy format of shape `HxWxC`
+
+    numpy.ndarray
+        Tranformed bounding box co-ordinates of the format `n x 4` where n is
+        number of bounding boxes and 4 represents `x1,y1,x2,y2` of the box
+
+    """
+
+    def __init__(self,p=0.5):
+        self.p=p
+        pass
+
+    def __call__(self, img, bboxes):
+        img_center = np.array(img.shape[:2])[::-1]/2
+        img_center = np.hstack((img_center, img_center))
+        if random.random() < self.p:
+            img = img[::-1, :, :]
+            bboxes[:, [1, 3]] += 2*(img_center[[1, 3]] - bboxes[:, [1, 3]])
+
+            box_h = abs(bboxes[:, 1] - bboxes[:, 3])
+
+            bboxes[:, 1] -= box_h
+            bboxes[:, 3] += box_h
+        return img, bboxes
+
 
 class RandomScale(object):
     """Randomly scales an image    
@@ -672,6 +773,56 @@ def __call__(self, img, bboxes):
 
         return img, bboxes
 
+class ShearY(object):
+    """Shears an image in Vertical direction   
+
+
+    Bounding boxes which have an area of less than 25% in the remaining in the 
+    transformed image is dropped. The resolution is maintained, and the remaining
+    area if any is filled by black color.
+
+    Parameters
+    ----------
+    shear_factor: float
+        Factor by which the image is sheared in the x-direction
+
+    Returns
+    -------
+
+    numpy.ndaaray
+        Sheared image in the numpy format of shape `HxWxC`
+
+    numpy.ndarray
+        Tranformed bounding box co-ordinates of the format `n x 4` where n is 
+        number of bounding boxes and 4 represents `x1,y1,x2,y2` of the box
+
+    """
+
+    def __init__(self, shear_factor = 0.2, vertical=False):
+        self.shear_factor = shear_factor
+
+
+    def __call__(self, img, bboxes):
+
+        shear_factor = self.shear_factor
+        if shear_factor < 0:
+            img, bboxes = VerticalFlip()(img, bboxes)
+
+        M = np.array([[1, 0, 0],[abs(shear_factor),1,0]])
+
+        nH =  img.shape[0] + abs(shear_factor*img.shape[1])
+
+        bboxes[:,[1,3]] += ((bboxes[:,[0,2]])*abs(shear_factor)).astype(int) 
+
+        img = cv2.warpAffine(img, M, (img.shape[1],int(nH)))
+
+        if shear_factor < 0:
+             img, bboxes = VerticalFlip()(img, bboxes)
+
+
+        return img, bboxes
+
+
 class Resize(object):
     """Resize the image in accordance to `image_letter_box` function in darknet