Skip to main content

deepsea is a package to segment and track single cells over the time-lapse phase contrast microscopy image sequences

Project description


This work presents a versatile and trainable deep-learning-based software, termed DeepSea, that allows for both segmentation and tracking of single cells in sequences of phase-contrast live microscopy images.


To download our datasets go to or:


  • Example of single cell image segmentation

from deepsea.test_single_image_segmentation import apply_img_segmentation
import cv2
import os

img = cv2.imread("segmentation_dataset/test/images/A11_z016_c001.png",0)
cv2.imwrite(os.path.join(output_dir, 'label_img.png'), label_img)
cv2.imwrite(os.path.join(output_dir, 'binary_mask.png'), binary_mask)
cv2.imwrite(os.path.join(output_dir, 'overlay_img.png'), overlay_img)
cv2.imwrite(os.path.join(output_dir, 'original_img_resized.png'), img)

  • Example of tracking the single set of cell image sequences

from deepsea.test_single_set_tracking import apply_cell_tracking
import cv2
import os

for img_name in sorted(os.listdir(single_image_set_dir)):

if tracked_imgs:
    for id, img in enumerate(tracked_imgs):
        cv2.imwrite(os.path.join(output_dir, 'img_{:04d}.png'.format(id)), img)

DeepSea GUI Software

Our DeepSea software is available on with examples and instructions. DeepSea software is a user-friendly and automated software designed to enable researchers to 1) load and explore their phase-contrast cell images in a high-contrast display, 2) detect and localize cell bodies using the pre-trained DeepSea segmentation model, 3) track and label cell lineages across the frame sequences using the pre-trained DeepSea tracking model, 4) manually correct the DeepSea models' outputs using user-friendly editing options, 5) train a new model with a new cell type dataset if needed, 6) save the results and cell label and feature reports on the local system. It employs our latest trained DeepSea models in the segmentation and tracking processes. It employs our last trained DeepSea models in the segmentation and tracking processes.

Useful Information

If you have any questions, contact us at

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

deepsea-0.0.3.tar.gz (15.4 MB view hashes)

Uploaded source

Built Distribution

deepsea-0.0.3-py3-none-any.whl (15.5 MB view hashes)

Uploaded py3

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