我训练了一个深度学习模型来检测圆，现在我绘制了矩形的边界框。这部分代码可以做到这一点。我正在使用python和opencv的组合。
> #initializing the lists detected bounding boxes, and class IDs, confidences respectively
boxes = []
classIDs = []
confidences = []
# for loop for each of the layer outputs
for out in raw_out_put:
# for loop for each of the detections
for detection in out:
# here we get the class ID and probability of the detected object inside the loop
scores = detection[5:]
classID = np.argmax(scores)
confidence = scores[classID]
# here we remove weak predictions by ensuring the detected probability is bigger than our threshold
if confidence > threshold_confidence:
# Here the bounding box coordinates are scaled back to the size of the image.
# having said that, YOLO returns the center-coordinates of the bounding box which is x, y and boxes' width and height
X = int(detection[0] * fWidth)
Y = int(detection[1] * fHeight)
width = int(detection[2] * fWidth)
height = int(detection[3] * fHeight)
#here by using x and y which is center coordinates, we calculate the top and left corner of the bounding box.
left = int(X - width / 2)
top = int(Y - height / 2)
# Then we update bounding box coordinates, confidences score, and class labels
classIDs.append(classID)
confidences.append(float(confidence))
boxes.append([left, top, width, height])
我的问题是，通过知道该圆的中心点（YOLO给出了边界框的中心点），我该如何以找到最低点的方式进行编码？ 说了我的图片中有多个圆圈，我需要找到所有圆圈的最低点。这一点将是圆的最低（向下）部分。图片可以说明我的需求。请查看以下链接。