# 旋转物体检测数据roLabelImg转DOTA格式

roLabelImg 工具仓库地址:https://github.com/cgvict/roLabelImg (opens new window)

# 1.进入可以画旋转检测框的模式

# 2.标注文件样式

roLabelImg标注旋转检测框时是先画一个常规的矩形框，然后绕矩形的中心点顺时针和逆时针旋转一定的角度来实现的。标注文件中对旋转检测框的定义是使用(cx, cy, width, height, angle)的格式定义的，如下：

<robndbox>
    <cx>1178.4388</cx>
    <cy>1004.6478</cy>
    <w>319.635</w>
    <h>273.2016</h>
    <angle>0.46</angle>
</robndbox>

(cx, cy)是旋转框的中心点像素坐标，w的定义是在roLabelImg中画初始矩形框时在图像x方向上的边长, 另一条边是h，画好初始矩形后，无论后续如何旋转，w和h所指的边都不会变。angle角的定义是旋转矩形检测框的w边和X轴正方向顺时针所成的角度，其大小为[0,pi)

初始矩形检测框：

调整姿态后检测框：

其中的大小为2.541593。

# 3.DOTA数据格式

DOTA是武汉大学开源的旋转物体检测数据集，其主页见https://captain-whu.github.io/DOTA/dataset.html (opens new window)。DOTA标注文件的格式为：

x1, y1, x2, y2, x3, y3, x4, y4, category, difficult

(x1, y1, x2, y2, x3, y3, x4, y4)分别是旋转物体检测框的四个顶点的坐标，category是检测框物体对象的类别

# 4.roLabelImg标注文件转`DOTA`格式

大多数旋转物体检测的开源算法的数据处理部分都支持DOTA格式，如商汤开源的mmrotate (opens new window),为了更快的在自己数据集上验证算法的有效性，最方便的算法就是将roLabelImg标注的xml文件转成上述的标签格式，roLabelImg标注文件转DOTA可分成四种情况。

1），且中心点C落在点1右侧
2），且中心点C落在点1左侧
3），且中心点C落在点1左侧
4），且中心点C落在点1右侧

A(x1, y1)，B(x3, y3)，D(x2, y2)，E(x4, y4)点的坐标由上述三角形之间的关系可以求得：

β = ∠ C A V_{2} = a r c t a n \frac{h}{w} + π - θ

d = \frac{\sqrt{w^{6} 2 + h^{2}}}{2}

x 1 = c x - d c o s β y 1 = c y + d s i n β x 2 = c x + d c o s β y 2 = c y - d s i n β x 3 = x 1 - h c o s (θ - \frac{π}{2}) y 4 = y 1 - h s i n (θ - \frac{π}{2}) x 4 = x 2 + h c o s (θ - \frac{π}{2}) y 4 = y 2 + h s i n (θ - \frac{π}{2})

同样可以求其他三种情况。

转换代码见：

def convert_rolabelimg2dota(xml_path:str) -> None:
"""
Args: 
    - `xml_path` (str) : path to roLabelImg label file, like /xx/xx.xml
    
Returns: 
    - `box_points` (list): shape (N, 8 + 1), N is the number of objects, 8 + 1 is \
        `(x1, y1, x2, y2, x3, y3, x4, y4, class_name)`
"""

with open(xml_path) as f:
    tree = ET.parse(f)
    root = tree.getroot()
    size = root.find('size')
    width = int(size.find('width').text)
    height = int(size.find('height').text)
    objects = root.iter('object')
    boxes = [] # list of tuple(cz, cy, w, h, angle), angle is in [0-pi)
    for obj in objects:
        if obj.find('type').text == 'robndbox':
            rbox_node = obj.find('robndbox')
            cat = obj.find('name').text
            rbox = dict()
            for key in ['cx', 'cy', 'w', 'h', 'angle']:
                    rbox[key] = float(rbox_node.find(key).text)
            boxes.append(list((*rbox.values(), cat)))
    print(f"bboxes: {boxes}")
    
    box_points = [] # list of box defined with four vertices
    for box in boxes:
        cx, cy, w, h, ag, cat = box
        alpha_w = math.atan(w / h)
        alpha_h = math.atan(h / w)
        d = math.sqrt(w**2 + h**2) / 2 
        if ag > math.pi / 2:
            beta = ag - math.pi / 2 + alpha_w
            if beta <= math.pi / 2:
                    x1, y1 = cx + d * math.cos(beta), cy + d * math.sin(beta)
                    x2, y2 = cx - d * math.cos(beta), cy - d * math.sin(beta)
            elif beta > math.pi / 2:
                    beta = math.pi - beta
                    x1, y1 = cx - d * math.cos(beta), cy + d * math.sin(beta)
                    x2, y2 = cx + d * math.cos(beta), cy - d * math.sin(beta)
            x3, y3 = x1 - h * math.cos(ag - math.pi / 2), y1 - h * math.sin(ag - math.pi / 2)
            x4, y4 = x2 + h * math.cos(ag - math.pi / 2), y2 + h * math.sin(ag - math.pi / 2) 
        elif ag <= math.pi / 2:
            beta = ag + alpha_h
            if beta <= math.pi / 2:
                    x1, y1 = cx + d * math.cos(beta), cy + d * math.sin(beta)
                    x2, y2 = cx - d * math.cos(beta), cy - d * math.sin(beta)
            elif beta > math.pi / 2:
                    beta = math.pi - beta
                    x1, y1 = cx - d * math.cos(beta), cy + d * math.sin(beta)
                    x2, y2 = cx + d * math.cos(beta), cy - d * math.sin(beta)
            x3, y3 = x1 - w * math.cos(ag), y1 - w * math.sin(ag)
            x4, y4 = x2 + w * math.cos(ag), y2 + w * math.sin(ag)
            points = np.array([x1, y1, x3, y3, x2, y2, x4, y4], dtype=np.int32)
            points[0::2] = np.clip(points[0::2], 0, width)
            points[1::2] = np.clip(points[1::2], 0, height)
        box_points.append([*points, cat])
    return box_points

完整代码见gitee仓库object_detection_task (opens new window)

# 参考资料

1.https://github.com/cgvict/roLabelImg (opens new window)

2.DOTA DATASET (opens new window)

← Modules of YOLOv5 ROI Pooling 与 ROI Align →