ilou89 / multiexposure_image_fusion Goto Github PK

View Code? Open in Web Editor NEW

2.0 2.0 1.0 1.45 MB

Implementation of the fabemd based exposure fusion algorithm in Qt/C++

QMake 3.32% C++ 96.68%

c-plus-plus empirical-mode-decomposition exposure-fusion image-fusion qt

multiexposure_image_fusion's People

Contributors

Stargazers

Watchers

Forkers

chnanan

multiexposure_image_fusion's Issues

Continuous Integration

Connect project to Appveyor (Windows) and later on to TravisCI (Linux, MacOS)

Design patterns

MainWindow must be singleton. Create MatrixFactory class to facilitate the creation of future derived classes. Consider other patterns as well.

-Smoke test (project build is performed without any problem)
-Coverage test (use either Qt Test or Google Test to achieve 100% code coverage)
-Functional test (fusion is performed correctly. If one image is used as input, the output should match the input)

Separable Max/Min filters

Optimize the max-min filters in Matrix2D class by performing them in two passes (row and column-wise) as the mean filter. Use an intermediate matrix between the two passes

Memory Leaks

Objects are currently created using "new" but deleting many of them was ignored. If it makes sense, replace them with smart pointers otherwise perform memory free manually.