Python utils.extract_bboxes() Examples

def load_image_gt(mask, config, anchors):
    """Generate the ground truth data for a mask.
    mask: [D, H, W]
    Returns:
    image: [1, D, H, W]
    class_ids: [instance_count] Integer class IDs
    bbox: [instance_count, (z1, y1, x1, z2, y2, x2)]
    mask: [num_classes, D, H, W]
    rpn_match: [batch, N] Integer (1=positive anchor, -1=negative, 0=neutral)
    rpn_bbox: [batch, N, (dz, dy, dx, log(dd), log(dh), log(dw))] Anchor bbox deltas
    """
    # Bounding boxes: [num_instances, (z1, y1, x1, z2, y2, x2)]
    bbox = utils.extract_bboxes(mask)  # we here use the whole liver + tumor as the gt-bbox
    bbox = utils.extend_bbox(bbox, mask.shape)  # extend the gt_bbox with 5% ratio in each dimension
    bbox = np.tile(bbox, (config.NUM_CLASSES - 1, 1))  # [num_classes - 1, (z1, y1, x1, z2, y2, x2)]

    # RPN Targets
    rpn_match, rpn_bbox = build_rpn_targets(anchors, np.array([bbox[0]]), config)

    # Add to batch
    rpn_match = rpn_match[:, np.newaxis]

    return rpn_match, rpn_bbox, bbox 

def visualize_instance_segmentation(data_base_dir, dataset_type, image_id, save_path='', verbose=True):
    split_dataset = SketchDataset(data_base_dir)
    split_dataset.load_sketches(dataset_type)
    split_dataset.prepare()

    original_image = split_dataset.load_image(image_id - 1)
    gt_mask, gt_class_id = split_dataset.load_mask(image_id - 1)
    gt_bbox = utils.extract_bboxes(gt_mask)

    if verbose:
        log('original_image', original_image)
        log('gt_class_id', gt_class_id)
        log('gt_bbox', gt_bbox)
        log('gt_mask', gt_mask)

    visualize.display_instances(original_image, gt_bbox, gt_mask, gt_class_id,
                                split_dataset.class_names, save_path=save_path) 

def load_image_gt(dataset, config, image_id, augment=False,
                  use_mini_mask=False):
    """Load and return ground truth data for an image (image, mask, bounding boxes).

    augment: If true, apply random image augmentation. Currently, only
        horizontal flipping is offered.
    use_mini_mask: If False, returns full-size masks that are the same height
        and width as the original image. These can be big, for example
        1024x1024x100 (for 100 instances). Mini masks are smaller, typically,
        224x224 and are generated by extracting the bounding box of the
        object and resizing it to MINI_MASK_SHAPE.

    Returns:
    image: [height, width, 3]
    shape: the original shape of the image before resizing and cropping.
    class_ids: [instance_count] Integer class IDs
    bbox: [instance_count, (y1, x1, y2, x2)]
    mask: [height, width, instance_count]. The height and width are those
        of the image unless use_mini_mask is True, in which case they are
        defined in MINI_MASK_SHAPE.
    """
    # Load image and mask
    image = dataset.load_image(image_id)
    mask, class_ids = dataset.load_mask(image_id)
    shape = image.shape
    image, window, scale, padding = utils.resize_image(
        image,
        min_dim=config.IMAGE_MIN_DIM,
        max_dim=config.IMAGE_MAX_DIM,
        padding=config.IMAGE_PADDING)
    mask = utils.resize_mask(mask, scale, padding)

    # Random horizontal flips.
    if augment:
        if random.randint(0, 1):
            image = np.fliplr(image)
            mask = np.fliplr(mask)

    # Bounding boxes. Note that some boxes might be all zeros
    # if the corresponding mask got cropped out.
    # bbox: [num_instances, (y1, x1, y2, x2)]
    bbox = utils.extract_bboxes(mask)

    # Active classes
    # Different datasets have different classes, so track the
    # classes supported in the dataset of this image.
    active_class_ids = np.zeros([dataset.num_classes], dtype=np.int32)
    source_class_ids = dataset.source_class_ids[dataset.image_info[image_id]["source"]]
    active_class_ids[source_class_ids] = 1

    # Resize masks to smaller size to reduce memory usage
    if use_mini_mask:
        mask = utils.minimize_mask(bbox, mask, config.MINI_MASK_SHAPE)

    # Image meta data
    image_meta = compose_image_meta(image_id, shape, window, active_class_ids)

    return image, image_meta, class_ids, bbox, mask 

def load_image_gt(dataset, config, image_id, augment=False,
                  use_mini_mask=False):
    """Load and return ground truth data for an image (image, mask, bounding boxes).

    augment: If true, apply random image augmentation. Currently, only
        horizontal flipping is offered.
    use_mini_mask: If False, returns full-size masks that are the same height
        and width as the original image. These can be big, for example
        1024x1024x100 (for 100 instances). Mini masks are smaller, typically,
        224x224 and are generated by extracting the bounding box of the
        object and resizing it to MINI_MASK_SHAPE.

    Returns:
    image: [height, width, 3]
    shape: the original shape of the image before resizing and cropping.
    class_ids: [instance_count] Integer class IDs
    bbox: [instance_count, (y1, x1, y2, x2)]
    mask: [height, width, instance_count]. The height and width are those
        of the image unless use_mini_mask is True, in which case they are
        defined in MINI_MASK_SHAPE.
    """
    # Load image and mask
    image = dataset.load_image(image_id)
    mask, class_ids = dataset.load_mask(image_id)
    shape = image.shape
    image, window, scale, padding = utils.resize_image(
        image,
        min_dim=config.IMAGE_MIN_DIM,
        max_dim=config.IMAGE_MAX_DIM,
        padding=config.IMAGE_PADDING)
    mask = utils.resize_mask(mask, scale, padding)

    # Random horizontal flips.
    if augment:
        if random.randint(0, 1):
            image = np.fliplr(image)
            mask = np.fliplr(mask)

    # Bounding boxes. Note that some boxes might be all zeros
    # if the corresponding mask got cropped out.
    # bbox: [num_instances, (y1, x1, y2, x2)]
    bbox = utils.extract_bboxes(mask)

    # Active classes
    # Different datasets have different classes, so track the
    # classes supported in the dataset of this image.
    active_class_ids = np.zeros([dataset.num_classes], dtype=np.int32)
    source_class_ids = dataset.source_class_ids[dataset.image_info[image_id]["source"]]
    active_class_ids[source_class_ids] = 1

    # Resize masks to smaller size to reduce memory usage
    if use_mini_mask:
        mask = utils.minimize_mask(bbox, mask, config.MINI_MASK_SHAPE)

    # Image meta data
    image_meta = compose_image_meta(image_id, shape, window, active_class_ids)

    return image, image_meta, class_ids, bbox, mask 

def load_image_gt(dataset, config, image_id, augment=False,
                  use_mini_mask=False):
    """Load and return ground truth data for an image (image, mask, bounding boxes).

    augment: If true, apply random image augmentation. Currently, only
        horizontal flipping is offered.
    use_mini_mask: If False, returns full-size masks that are the same height
        and width as the original image. These can be big, for example
        1024x1024x100 (for 100 instances). Mini masks are smaller, typically,
        224x224 and are generated by extracting the bounding box of the
        object and resizing it to MINI_MASK_SHAPE.

    Returns:
    image: [height, width, 3]
    shape: the original shape of the image before resizing and cropping.
    class_ids: [instance_count] Integer class IDs
    bbox: [instance_count, (y1, x1, y2, x2)]
    mask: [height, width, instance_count]. The height and width are those
        of the image unless use_mini_mask is True, in which case they are
        defined in MINI_MASK_SHAPE.
    """
    # Load image and mask
    image = dataset.load_image(image_id)
    mask, class_ids = dataset.load_mask(image_id)
    shape = image.shape
    image, window, scale, padding = utils.resize_image(
        image,
        min_dim=config.IMAGE_MIN_DIM,
        max_dim=config.IMAGE_MAX_DIM,
        padding=config.IMAGE_PADDING)
    mask = utils.resize_mask(mask, scale, padding)

    # Random horizontal flips.
    if augment:
        if random.randint(0, 1):
            image = np.fliplr(image)
            mask = np.fliplr(mask)

    # Bounding boxes. Note that some boxes might be all zeros
    # if the corresponding mask got cropped out.
    # bbox: [num_instances, (y1, x1, y2, x2)]
    bbox = utils.extract_bboxes(mask)

    # Active classes
    # Different datasets have different classes, so track the
    # classes supported in the dataset of this image.
    active_class_ids = np.zeros([dataset.num_classes], dtype=np.int32)
    source_class_ids = dataset.source_class_ids[dataset.image_info[image_id]["source"]]
    active_class_ids[source_class_ids] = 1

    # Resize masks to smaller size to reduce memory usage
    if use_mini_mask:
        mask = utils.minimize_mask(bbox, mask, config.MINI_MASK_SHAPE)

    # Image meta data
    image_meta = compose_image_meta(image_id, shape, window, active_class_ids)

    return image, image_meta, class_ids, bbox, mask 

def load_image_gt(dataset, config, image_id, augment=False,
                  use_mini_mask=False):
    """Load and return ground truth data for an image (image, mask, bounding boxes).

    augment: If true, apply random image augmentation. Currently, only
        horizontal flipping is offered.
    use_mini_mask: If False, returns full-size masks that are the same height
        and width as the original image. These can be big, for example
        1024x1024x100 (for 100 instances). Mini masks are smaller, typically,
        224x224 and are generated by extracting the bounding box of the
        object and resizing it to MINI_MASK_SHAPE.

    Returns:
    image: [height, width, 3]
    shape: the original shape of the image before resizing and cropping.
    class_ids: [instance_count] Integer class IDs
    bbox: [instance_count, (y1, x1, y2, x2)]
    mask: [height, width, instance_count]. The height and width are those
        of the image unless use_mini_mask is True, in which case they are
        defined in MINI_MASK_SHAPE.
    """
    # Load image and mask
    image = dataset.load_image(image_id)
    mask, class_ids = dataset.load_mask(image_id)
    shape = image.shape
    image_resize, window, scale, padding = utils.resize_image(
        image,
        min_dim=config.IMAGE_MIN_DIM,
        max_dim=config.IMAGE_MAX_DIM,
        padding=config.IMAGE_PADDING)
    # mask = utils.resize_mask(mask, scale, padding)
    mask = mask

    # Random horizontal flips.
    if augment:
        if random.randint(0, 1):
            image = np.fliplr(image)
            mask = np.fliplr(mask)

    # Bounding boxes. Note that some boxes might be all zeros
    # if the corresponding mask got cropped out.
    # bbox: [num_instances, (y1, x1, y2, x2)]
    bbox = utils.extract_bboxes(mask)

    # Active classes
    # Different datasets have different classes, so track the
    # classes supported in the dataset of this image.
    active_class_ids = np.zeros([dataset.num_classes], dtype=np.int32)
    source_class_ids = dataset.source_class_ids[dataset.image_info[image_id]["source"]]
    active_class_ids[source_class_ids] = 1

    # Resize masks to smaller size to reduce memory usage
    if use_mini_mask:
        mask = utils.minimize_mask(bbox, mask, config.MINI_MASK_SHAPE)

    # Image meta data
    image_meta = compose_image_meta(image_id, shape, window, active_class_ids)

    return image, image_meta, class_ids, bbox, mask 

def load_image_gt(dataset, config, image_id, augment=True,
                  use_mini_mask=False):
    """Load and return ground truth data for an image (image, mask, bounding boxes).

    augment: If true, apply random image augmentation. Currently, only
        horizontal flipping is offered.
    use_mini_mask: If False, returns full-size masks that are the same height
        and width as the original image. These can be big, for example
        1024x1024x100 (for 100 instances). Mini masks are smaller, typically,
        224x224 and are generated by extracting the bounding box of the
        object and resizing it to MINI_MASK_SHAPE.

    Returns:
    image: [height, width, 3]
    shape: the original shape of the image before resizing and cropping.
    class_ids: [instance_count] Integer class IDs
    bbox: [instance_count, (y1, x1, y2, x2)]
    mask: [height, width, instance_count]. The height and width are those
        of the image unless use_mini_mask is True, in which case they are
        defined in MINI_MASK_SHAPE.
    """
    # Load image and mask
    image = dataset.load_image(image_id)
    mask, class_ids = dataset.load_mask(image_id)
    shape = image.shape
    image_resize, window, scale, padding = utils.resize_image(
        image,
        min_dim=config.IMAGE_MIN_DIM,
        max_dim=config.IMAGE_MAX_DIM,
        padding=config.IMAGE_PADDING)
    # mask = utils.resize_mask(mask, scale, padding)
    mask = mask

    # Random horizontal flips.
    if augment:
        if random.randint(0, 1):
            image = np.fliplr(image)
            mask = np.fliplr(mask)

    # Bounding boxes. Note that some boxes might be all zeros
    # if the corresponding mask got cropped out.
    # bbox: [num_instances, (y1, x1, y2, x2)]
    bbox = utils.extract_bboxes(mask)

    # Active classes
    # Different datasets have different classes, so track the
    # classes supported in the dataset of this image.
    active_class_ids = np.zeros([dataset.num_classes], dtype=np.int32)
    source_class_ids = dataset.source_class_ids[dataset.image_info[image_id]["source"]]
    active_class_ids[source_class_ids] = 1

    # Resize masks to smaller size to reduce memory usage
    if use_mini_mask:
        mask = utils.minimize_mask(bbox, mask, config.MINI_MASK_SHAPE)

    # Image meta data
    image_meta = compose_image_meta(image_id, shape, window, active_class_ids)

    return image, image_meta, class_ids, bbox, mask