ImageNormalization

Version 0.1

What's Included

ImageNormalization includes scripts that normalize images from a Quadband sensor based on exposure time and the analog and digital gain sensors.

Installation Instructions

These instructions have only been tested on a 64-bit Windows 10 machine in both Python 2.7 and 3.6 environments. and #### Package Installation

1. Inside of a terminal, cd to ImageNormalization/
2. Run python steup.py install

Now the multispectral package is available to use in Python.

Anaconda (Optional)

For more information on using Anaconda, the reader is encouraged to visit https://docs.continuum.io/anaconda/navigator/tutorials/manage-environments

1. Download and install Anaconda (Python 3.6 version recommended)
2. Launch Anaconda Navigator
3. Create a new Python 3.6 environment with a relevant name (e.g. "env-multispectral")

Dependencies

Currently the only dependancy of ImageNormalization is the Pillow image processing library. However, the image manipulation may be changed to use the gdal library in the future as other internal tools, such as those in AnalysisTools, use that for manipulation.

Using MPP

Currently, there is only one script included with the ImageNormalization package. In order to work correctly, each image directory should contain the pix4d.csv and autoexposure.csv files.

Command Line Tools

To calibrate one or more directories of images using the ImageNormalization command line tools, use:

python <path-to-dir>/quadband.py <image-directory-1> <image-directory-2> ... <image-directory-N>

Python Functions

To use the ImageNormalization tools in a Python script, import the package using import multispectral. Each module is specified below.

• quadband module:
  • post_process([list-of-image-directories], processed_subdir='processed')