Skip to main content

Vision features of generalistic use

Project description

PyPI - Python Version PyPI version

UIB - V Features is a library to extract useful features of different types(morphological, texture and color). To increase the flexibility of the library all functions can be used with masks or contours.

UIB - V Features was originally developed by researchers of UGIVIA. This research group is centered on computer vision and artificial intelligence.

All the features available are grouped in one iterator to simplify the use of this library.


Install the library is very simple with pip

$ pip install uib-vfeatures

List of features


  • Solidity
  • Convex hull perimeter
  • Convex hull area
  • Bounding box area
  • Rectangularity
  • Minor radius
  • Maximum radius
  • Feret
  • Breadh
  • Circularity
  • Roundness
  • Feret Angle
  • Eccenctricity
  • Center
  • Sphericity
  • Aspect Ratio
  • Area equivalent diameter
  • Perimeter equivalent diameter
  • Equivalent elipse area
  • Compactness
  • Area
  • Convexity
  • Shape
  • Perimeter


  • Mean of the LAB channels
  • Mean of the RGB channels
  • Mean of the HSV channels
  • Standard deviation of the LAB channels
  • Standard deviation of the RGB channels
  • Standard deviation of the HSV channels

Texture features

  • Contrast
  • Dissimilarity
  • Homogeneity
  • ASN
  • Energy
  • Correlation


We're going to use our library with a mask image .

from uib_vfeatures.masks import Masks
from uib_vfeatures import Features_mask as ftrs
import cv2

First of all we read the image from a file, then we try our features with visualizations. We only have three features with visualization: the bounding box area, the eccentricity and the solidity.

mask = cv2.imread("mask.jpg")

Masks.bounding_box_area(mask, True)

Masks.eccentricity(mask, True)
Masks.solidity(mask, True)


You can use an iterator and implement every morpholical feature.

features = {}

for key, func in features.items():
    features[key] = func(mask)

As a result we had a dicctionary of the form {'Feature_name': value}


If you use this code, please cite

    title = {Sickle-cell disease diagnosis support selecting the most appropriate machine learning method: Towards a general and interpretable approach for cell morphology analysis from microscopy images},
    author = {Nataša Petrović and Gabriel Moyà-Alcover and Antoni Jaume-i-Capó and Manuel González-Hidalgo},
    journal = {Computers in Biology and Medicine},
    volume = {126},
    pages = {104027},
    year = {2020},
    issn = {0010-4825},
    doi = {},
    url = {}

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

uib_vfeatures-0.7.tar.gz (11.4 kB view hashes)

Uploaded source

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