Dependencies can be build from sources or obtained from package managers when available.
NOTE : Dockerfile and docker-dev-init may provide some useful info about how to build those dependencies,
but we rather recommend to follow each dependency installation instructions.
mkdir build && cd build
ccmake ../ (or cmake-gui)
make
NOTE : Only a handful of OpenCV functions are used by the pyramid building algorithm. Linking statically to OpenCV is recommended for container distribution in order to reduce the size of the container.
Docker distribution available on DockerHub at https://hub.docker.com/r/wipp/pyramid-building
A Docker image can also be build from source.
To build the Docker image for development/testing, run:
docker build -t pyramid-building:dev .
You can then run a simple test for building a pyramid using the Docker image:
sh run-simple-test.sh
Example:
./main
-v "/path/to/stitching_vector/img-global-positions-1.txt"
-i "/path/to/images/"
-o "/path/to/outputDir"
-t 256 -d 8U -b "overlay"
-n "pyramidName"
Environment variables can be set to generate logs.
GLOG_logtostderr=1; #print tostderr instead of log file GLOG_v=[0..4] #set log level between 0 and 4 for more detailed logs
Log Levels :
0/ no logs. 1/ basic info and execution time. 2/ algorithmic steps. 3/ detailed logs. 4/ very detailed logs.
A basic integration of the deepzoom js library is provided to check the result of a pyramid building job. A http server can quickly be deployed with npm.
sudo apt-get install npm
sudo npm install -g http-server
Edit test-webapp/index.html and update the tileSources map entry with the path to your pyramid dzi file.
(Make sure to provide a relative path to the test-webapp directory).
Start the http-server (the root must allow access the pyramid directory).
http-server ..
Visit http://127.0.0.1:8080