Nick Peihl nickpeihl

Pansharpening notes, mid-2021

First posted in August 2021. This is basically a snapshot of my thinking about pansharpening at that time; I’m not making any substantial updates. Last typo and clarity fixes in February 2023.

Preface

This is a collection of notes on how I’ve been approaching convolutional neural networks for pansharpening. It’s an edited version of an e-mail to a friend who had asked about this tweet, so it’s informal and somewhat silly; it’s not as polished as, say, a blog post would be. It’s basically the advice I would give to an image processing hobbyist before they started working on pansharpening.

If you want a more serious introduction, start with the literature review in Learning deep multiresolution representations for pansharpening. Most of the academic work I would recommend is mentioned there.

How to submit Kibana Vega question

It is usually very difficult to debug Vega questions without having your data. To make it easier, please follow these steps to include data with your graph when posting:

Reduce your data query to the smallest possible dataset, e.g. set the time range to 15 minutes. It will work as long as it is not empty and represents your data well enough.
Open Browser Debugger (for Chrome, right click and click Inspect)
Switch to the Console tab
Copy the right command, paste it in the console at the > symbol and hit enter (check the schema in your graph to see if you use Vega or Vega-Lite)

cat data.geojson | jq -c '.features | .[]' > outfile.jsonld

Technocratic Oath

I swear to fulfill, to the best of my ability and judgment, this covenant:

I will respect the hard-won scientific gains of those ~~physicians~~ in whose steps I walk, and gladly share such knowledge as is mine with those who are to follow.

I will apply, for the benefit of the ~~sick~~ user, all measures which are required, avoiding those twin traps of overtreatment and therapeutic nihilism.

I will remember that there is art to ~~medicine~~ software as well as science, and that warmth, sympathy, and understanding may outweigh the ~~surgeon's knife or the chemist's drug~~ need for another dependency.

	Mute these words in your settings here: https://twitter.com/settings/muted_keywords

	ActivityTweet
	generic_activity_highlights
	generic_activity_momentsbreaking
	RankedOrganicTweet
	suggest_activity
	suggest_activity_feed
	suggest_activity_highlights
	suggest_activity_tweet

	{
	$schema: https://vega.github.io/schema/vega/v3.json
	padding: 10
	config: {
	kibana: {tooltips: false}
	}
	signals: [
	{
	name: rotation
	value: 65

	# Description: Boxstarter Script
	# Author: Jess Frazelle <jess@linux.com>
	# Last Updated: 2017-09-11
	#
	# Install boxstarter:
	# . { iwr -useb http://boxstarter.org/bootstrapper.ps1 } \| iex; get-boxstarter -Force
	#
	# You might need to set: Set-ExecutionPolicy RemoteSigned
	#
	# Run this boxstarter by calling the following from an elevated command-prompt:

	var socket = dgram.createSocket({ type: 'udp4', reuseAddr: true })
	socket.bind(5004, function () {
	socket.addMembership('239.255.42.42', getIpForInterface('en2'))
	socket.setMulticastTTL(255)
	})

	socket.on('message', function (m) {
	// m is a buffer
	})

	"""
	First, this was written back when I wrote Python a bit more like C - I know it needs some work! But for setting delimiters,
	you can do the following:

	my_dataset = data_file(r"path_to_data")
	my_dataset.delim_open # gives you the opening delimeter
	my_dataset.delim_close # gives you the closing delimeter
	"""

	import logging

	# python pca_multiband.py input.jpeg output.tif
	# n-band image -> PCA -> n-band TIFF image
	# with lots of hackety assumptions
	# (e.g., output is same type as input)

	from sys import argv
	import rasterio as rio
	import numpy as np
	from sklearn import decomposition