Operator ID: daft.las.functions.image.image_blackborder_crop.ImageBlackBorderCrop

Description #

Image black border detection and cropping processor, supports multiple detection algorithms and output formats.

Key features #

• Four professional-grade black border detection algorithms:
  • Threshold detection (threshold) – fastest speed, suitable for pure black borders or solid color borders
  • Edge detection (edge) – based on Canny edge detection, suitable for blurred or gradient black borders
  • Histogram detection (histogram) – based on pixel distribution, suitable for low-contrast black borders
  • Automatic detection (auto) – intelligently selects the optimal algorithm (recommended)
• Multi-format input support:
  • URL address (image_url)
  • Base64 encoding (image_base64)
  • Binary stream (image_binary)
• Multi-format output support:
  • JPEG/JPG: Automatically converts RGBA to RGB, preserves DPI
  • PNG: Preserves transparency channel, lossless saving
  • WebP: Supports transparency channel, flexible saving
• Dual output modes:
  • Direct Base64 encoding output
  • TOS/local persistent storage
• Robustness assurance:
  • Minimum black border size filtering (prevents excessive cropping)
  • Boundary validation (prevents cropping beyond image boundaries)
  • Automatic fallback to original image if detection fails

Operator parameters #

Input #

Output #

An array of dictionaries containing processing results; each element includes:

• base64: Base64 encoding of the cropped image;
• image_path: Local/TOS storage path (effective when output directory is configured)

Parameters #

If a parameter does not have a default value, it is required.

Examples #

The following code demonstrates how to use daft to run the operator to crop black borders from images.

from __future__ import annotations

import os

import daft
from daft import col
from daft.las.functions.image.image_blackborder_crop import ImageBlackBorderCrop
from daft.las.functions.udf import las_udf

if __name__ == "__main__":
    # Content will be saved to the specified TOS path. Therefore, you need to set environment variables to ensure you have permission to write to TOS, including: ACCESS_KEY, SECRET_KEY, TOS_ENDPOINT, TOS_REGION, TOS_TEST_DIR
    TOS_DIR = os.getenv("TOS_TEST_DIR", "tos_bucket")
    output_tos_dir = f"tos://{TOS_DIR}/image/image_blackborder_crop"

    if os.getenv("DAFT_RUNNER", "native") == "ray":
        import logging

        import ray

        def configure_logging():
            logging.basicConfig(
                level=logging.INFO,
                format="%(asctime)s - %(name)s - %(levelname)s - %(message)s",
                datefmt="%Y-%m-%d %H:%M:%S.%s".format(),
            )
            logging.getLogger("tracing.span").setLevel(logging.WARNING)
            logging.getLogger("daft_io.stats").setLevel(logging.WARNING)
            logging.getLogger("DaftStatisticsManager").setLevel(logging.WARNING)
            logging.getLogger("DaftFlotillaScheduler").setLevel(logging.WARNING)
            logging.getLogger("DaftFlotillaDispatcher").setLevel(logging.WARNING)

        ray.init(dashboard_host="0.0.0.0", runtime_env={"worker_process_setup_hook": configure_logging})
        daft.set_runner_ray()

    daft.set_execution_config(actor_udf_ready_timeout=600)
    daft.set_execution_config(min_cpu_per_task=0)

    tos_dir_url = os.getenv("TOS_DIR_URL", "las-cn-beijing-public-online.tos-cn-beijing.volces.com")
    samples = {
        "image": [f"https://{tos_dir_url}/public/shared_image_dataset/test_blackborder.png"],
        "image_name": ["cat_with_border_crop"],
    }

    image_suffix = ".jpg"
    image_src_type = "image_url"
    detect_algorithm = "auto"
    black_threshold = 10
    edge_sensitivity = 1.0
    min_border_size = 1
    target_dpi = [72, 72]
    ds = daft.from_pydict(samples)
    ds = ds.with_column(
        "image_blackborder_crop",
        las_udf(
            ImageBlackBorderCrop,
            construct_args={
                "image_suffix": image_suffix,
                "output_dir": output_tos_dir,
                "image_src_type": image_src_type,
                "detect_algorithm": detect_algorithm,
                "black_threshold": black_threshold,
                "edge_sensitivity": edge_sensitivity,
                "min_border_size": min_border_size,
                "target_dpi": target_dpi,
            },
            batch_size=1,
        )(col("image"), col("image_name")),
    )

    ds.show()
# ╭────────────────────────────────┬──────────────────────┬────────────────────────────────────────────╮
# │ image                          ┆ image_name           ┆ image_blackborder_crop                     │
# │ ---                            ┆ ---                  ┆ ---                                        │
# │ String                         ┆ String               ┆ Struct[base64: String, image_path: String] │
# ╞════════════════════════════════╪══════════════════════╪════════════════════════════════════════════╡
# │ https://las-ai-qa-online.tos-… ┆ cat_with_border_crop ┆ {base64: /9j/4AAQSkZJRgABAQEA…             │
# ╰────────────────────────────────┴──────────────────────┴────────────────────────────────────────────╯