144c6c7c2a60531ed72afe07cc575c261be9ec2f,petastorm/codecs.py,CompressedImageCodec,decode,#CompressedImageCodec#Any#Any#,69

Before Change

    def decode(self, unischema_field, value):
        Decode the image using OpenCV
        numpy_image = cv2.imdecode(np.frombuffer(value, dtype=np.uint8), cv2.IMREAD_UNCHANGED)
        return numpy_image

    def spark_dtype(self):
        return BinaryType()

After Change

        // cv returns a BGR or grayscale image. Convert to RGB (unless a grayscale image).
        image_bgr_or_gray = cv2.imdecode(np.frombuffer(value, dtype=np.uint8), cv2.IMREAD_UNCHANGED)
        if len(image_bgr_or_gray.shape) == 2:
            // Greyscale image
            return image_bgr_or_gray
        elif len(image_bgr_or_gray.shape) == 3 and image_bgr_or_gray.shape[2] == 3:
            // Convert BGR to RGB (opencv assumes BGR)
            image_rgb = image_bgr_or_gray[:, :, (2, 1, 0)]
            return image_rgb
        else:
            raise ValueError("Unexpected image dimensions. Supported dimensions are (H, W) or (H, W, 3). "
                               "Got {}".format(image_bgr_or_gray.shape))

    def spark_dtype(self):
        return BinaryType()

Instances