Skip to main content

Helpful functions for Computer Vision tasks

Project description

🦌 Kano - Tools for Computer Vision Tasks

Kano CI

Kano is a Python package providing utility functions for Computer Vision tasks. Its primary focus is simplifying lengthy functions, allowing developers to concentrate more on the main processes.

📥 Installation

The latest released version is available on PyPI. You can install it by running the following command in your terminal:

pip install kano-cv

🚀 Usage

🗃️ YOLO Datasets Splitting/Merging

Test these utilities here: Open in Colab

If you are using Roboflow to label and struggling to merge many Workspaces or YOLO format projects, Kano provides a utility to merge them with just one command:

from kano.complex.roboflow import merge_datasets


# Each dataset folder contains two folders "images" and "labels"
folders = [
    "Dataset_1",
    "Dataset_2",
]

merge_datasets(folders, merged_folder_path="dataset")

You can also split a dataset into train/valid/test with your own split ratio with only one command:

from kano.complex.roboflow import split_dataset


# Split a dataset (contains "images" and "labels")
# into train, valid, test folders with a given ratio
split_dataset(
    dataset_path="dataset/train",
    train_percent=80,
    valid_percent=20,
    target_folder="splitted_dataset",
)

📁 Files/Folders Manipulating

Test these utilities here: Open in Colab

Kano is designed to run many common functions in just one line:

from kano.file_utils import create_folder, print_foldertree, remove_folder

# Create a folder and its subfolder without errors
create_folder("folder_A/subfolder")

for i in range(2):
    with open(f"folder_A/subfolder/file_{i}.txt", "w") as f:
        pass

print_foldertree("folder_A")
# folder_A (2 files)
# |-- subfolder (2 files)

# Remove a folder with its content without errors
remove_folder("folder_A/subfolder")

You can even zip many folders (and files) by providing their paths and the destination path in a function call:

zip_paths(["folder_A", "folder_B"], "zipfile.zip")

🖼️ Images Processing

Test these utilities here: Open in Colab

You can quickly download an image using a URL and show it in IPython notebooks or Python files:

from kano.image_utils import download_image, show_image


image = download_image("https://avatars.githubusercontent.com/u/77763935?v=4", "image.jpg")

# using a numpy array
show_image(image)

# using a file path
show_image("image.jpg")

or you can get a random image with a specific size:

from kano.image_utils import get_random_picture


image = get_random_picture(width=400, height=300, save_path="random_image.jpg")

# using a numpy array
show_image(image)

# using a file path
show_image("random_image.jpg")

🎞️ Videos Processing

Test these utilities here: Open in Colab

Kano helps you extract images from a video. For demo purposes, I will download a video from YouTube using pytube. If you find this function helpful, please give a star to the original repo.

from kano.video_utils import download_youtube_video, extract_frames


download_youtube_video("https://www.youtube.com/watch?v=<VIDEOID>", "video.mp4")

# Get 1 image per 2 seconds
extract_frames(
    video_path="video.mp4",
    target_folder="frames",
    seconds_interval=2,
)

🙋‍♂️ Contributing to Kano

All contributions, bug reports, bug fixes, enhancements, and ideas are welcome. Feel free to create pull requests or issues so that we can improve this library together.

🔑 License

Kano is licensed under the MIT license.

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

kano-cv-1.8.4.tar.gz (11.1 kB view details)

Uploaded Source

Built Distribution

kano_cv-1.8.4-py3-none-any.whl (10.2 kB view details)

Uploaded Python 3

File details

Details for the file kano-cv-1.8.4.tar.gz.

File metadata

  • Download URL: kano-cv-1.8.4.tar.gz
  • Upload date:
  • Size: 11.1 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? Yes
  • Uploaded via: twine/4.0.2 CPython/3.11.8

File hashes

Hashes for kano-cv-1.8.4.tar.gz
Algorithm Hash digest
SHA256 2804bdb55f8065c2f923166823dcc89f1408b04d63f26bc975c61afc17276880
MD5 82c0ad437d56327b75fb67bede75a54a
BLAKE2b-256 d955534144ee282705d38c196e4ac29b2ebc7b9a327df306e96a145665565485

See more details on using hashes here.

File details

Details for the file kano_cv-1.8.4-py3-none-any.whl.

File metadata

  • Download URL: kano_cv-1.8.4-py3-none-any.whl
  • Upload date:
  • Size: 10.2 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? Yes
  • Uploaded via: twine/4.0.2 CPython/3.11.8

File hashes

Hashes for kano_cv-1.8.4-py3-none-any.whl
Algorithm Hash digest
SHA256 f514ad2ad69d5fbd44eadbf0da703a37636f202a4bb8d0b4eb6d876012d38b84
MD5 aaa55e31f5c82f8a1da02f250c01b39f
BLAKE2b-256 b4f2c4909d40bb245ae017437915807a2731b045ed5b8d516cbe675eca331995

See more details on using hashes here.

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page