目标检测 YOLOv5 - 模型推理预处理 letterbox

flyfish

版本：YOLOv5 6.2
假如图片大小是1080 * 1920 （height * width ）
width = 1920
height = 1080

当模型输入是 640 * 640时
shapes = (1080, 1920), (0.33, 0.33), (0.0, 140.0)

640/ 1920 = 0.33
图片缩放到 360 * 640 （height * width ）

当前 模型 输入 必须 640 * 640 时
就将 360 * 640 的图片 装进 640 * 640中
多余的部分都是灰色边框
140是图片上下边框各自的高度

当模型输入 不必须 640 * 640 时
可以做成填充最小矩形灰色边框能被32整除
360 / 32 = 11.25
384 / 32 = 12
360 * 640 变成了 384 * 640

这样图片上下有很少的灰色边框
1080 * 1920 （height * width ）变 360 * 640 （height * width ）

def load_image(self, i):
    # Loads 1 image from dataset index 'i', returns (im, original hw, resized hw)
    im, f, fn = self.ims[i], self.im_files[i], self.npy_files[i],
    if im is None:  # not cached in RAM
        if fn.exists():  # load npy
            im = np.load(fn)
        else:  # read image
            im = cv2.imread(f)  # BGR
            assert im is not None, f'Image Not Found {f}'
        h0, w0 = im.shape[:2]  # orig hw
        r = self.img_size / max(h0, w0)  # ratio
        if r != 1:  # if sizes are not equal
            interp = cv2.INTER_LINEAR if (self.augment or r > 1) else cv2.INTER_AREA
            im = cv2.resize(im, (int(w0 * r), int(h0 * r)), interpolation=interp)
        return im, (h0, w0), im.shape[:2]  # im, hw_original, hw_resized
    return self.ims[i], self.im_hw0[i], self.im_hw[i]  # im, hw_original, hw_resized

360 * 640 （height * width ）变 384 * 640 （height * width ）

def letterbox(im, new_shape=(640, 640), color=(114, 114, 114), auto=True, scaleFill=False, scaleup=True, stride=32):
    # Resize and pad image while meeting stride-multiple constraints
    shape = im.shape[:2]  # current shape [height, width]
    if isinstance(new_shape, int):
        new_shape = (new_shape, new_shape)

    # Scale ratio (new / old)
    r = min(new_shape[0] / shape[0], new_shape[1] / shape[1])
    if not scaleup:  # only scale down, do not scale up (for better val mAP)
        r = min(r, 1.0)

    # Compute padding
    ratio = r, r  # width, height ratios
    new_unpad = int(round(shape[1] * r)), int(round(shape[0] * r))
    dw, dh = new_shape[1] - new_unpad[0], new_shape[0] - new_unpad[1]  # wh padding
    if auto:  # minimum rectangle
        dw, dh = np.mod(dw, stride), np.mod(dh, stride)  # wh padding
    elif scaleFill:  # stretch
        dw, dh = 0.0, 0.0
        new_unpad = (new_shape[1], new_shape[0])
        ratio = new_shape[1] / shape[1], new_shape[0] / shape[0]  # width, height ratios

    dw /= 2  # divide padding into 2 sides
    dh /= 2

    if shape[::-1] != new_unpad:  # resize
        im = cv2.resize(im, new_unpad, interpolation=cv2.INTER_LINEAR)
    top, bottom = int(round(dh - 0.1)), int(round(dh + 0.1))
    left, right = int(round(dw - 0.1)), int(round(dw + 0.1))
    im = cv2.copyMakeBorder(im, top, bottom, left, right, cv2.BORDER_CONSTANT, value=color)  # add border
    return im, ratio, (dw, dh)