updated readme

README.md

@@ -1,15 +1,32 @@
 # CLAHE3D
+
 A three dimensional contrast enhancement code written in `C++`. Works as standalone C++ application or through a MATLAB interface specified as a `*.mex` file. Since I am using it at the moment only through `mex` I did not implement any file import or export functions for the `C++` based version but can do that upon reasonable request. Histograms are calculated for subvolumes and afterwards interpolated to the currently adapted pixel through trilinear interpolation.
 
 # MATLAB
 Change to folder and run `mex -O clahe3dmex.cpp interpGrid.cpp histeq.cpp`. Afterwards you can use the function as:
-`clahe3dmex(vol, volSize, clipLimit, binSize);`
-where `vol` is the three dimensional volume of type `single`, `volSize` is the subvolume size specified as a 3 element vector of type `uint64`, `clipLimit` is the clipping threshold specified as `single`, and `binSize` is the number of histogram bins specified as `uint64`. 
+`clahe3dmex(interpVol, subVolSize, spacingSubVols, clipLimit, binSize);`
+where 
+
+*  `vol` is the three dimensional volume of type `single`,
+*  `subVolSize` is the container size used to create the histograms subvolume size specified as a 3 element vector of type `uint64`
+*  `spacingSubVols` is the spacing vector between the histograms specified as `uint64`
+*  `clipLimit` is the clipping threshold specified as `single`
+*  and `binSize` is the number of histogram bins specified as `uint64`. 
 
-Tested using `MATLAB 2019 A` and `MATLAB 2019 B`.
+Tested using `MATLAB 2019 A` and `MATLAB 2019 B`. The function `utest_bone.m` serves as an illustration.
 
 # C++
-`makefile` in main folder. Simply clone repository into your local filesystem and compile using `make main`. Tested on ArchLinux using `g++` compiler. Object files will be sotred in 'obj' directory, executable goes to `bin/main`.
+
+The code is compiled using `cmake`.
+
+```bash
+cd Debug
+cmake ..
+make all
+./main_exp
+```
+
+An example of how to use the main class can be found in `src/main.cpp`.
 
 # Feature request / bug report
 Feel free to contact me if you need the addition of a feature or if you want to report a bug. Next step of this project is going to be a `CUDA` capable parallelized version of the code.

src/clahe3dmex.cpp

@@ -1,3 +1,10 @@
+/*
+	File: clahe3dmex.cpp
+	Author: Urs Hofmann
+	Mail: mail@hofmannu.org
+	Date: 11.02.2022
+*/
+
 #include <fstream>
 #include "interpGrid.h"
 #include "histeq.h"
@@ -6,7 +13,6 @@
 using namespace std;
 
 // clahe3dmex(interpVol, subVolSize, spacingSubVols, clipLimit, binSize);
-
 void mexFunction(int nlhs, mxArray *plhs[], int nrhs, const mxArray *prhs[]){
 
 	float* inputVol = (float *) mxGetData(prhs[0]); // input volume
@@ -17,9 +23,7 @@ void mexFunction(int nlhs, mxArray *plhs[], int nrhs, const mxArray *prhs[]){
 	uint64_t * subVolSpacing = (uint64_t*) mxGetData(prhs[2]); // size of bins [z, x, y]
 	float* clipLimit = (float *) mxGetData(prhs[3]);
 	uint64_t* binSize = (uint64_t*) mxGetData(prhs[4]);
-
 	uint64_t volumeSize[3] = {nZ, nX, nY};
-
 
 	printf("[clahe3d] initializing histogram handler\n");
 	histeq histHandler;

src/main.cpp

@@ -32,9 +32,9 @@ int main(){
 
 	// initialize some parameters
 	const float clipLimit = 0.1;
-	const uint64_t binSize = 50;
+	const uint64_t binSize = 250;
 	const uint64_t subVolSize[3] = {31, 31, 31};
-	const uint64_t subVolSpacing[3] = {10, 10, 10};
+	const uint64_t subVolSpacing[3] = {20, 20, 20};
 	const uint64_t gridSize[3] = {nZ, nX, nY};
 
 	histeq histHandler;

test.m → utest_bone.m

@@ -27,8 +27,6 @@
 	error("Function returned invalid values");
 end
 
-% vtkwrite('/media/hofmannu/hofmannu/clahed.vtk', 'structured_points', 'clahed', saftedVol, 'spacing', Settings.z(2) - Settings.z(1), Settings.dX, Settings.dY, 'BINARY');
-
 newSlice = interpVol(100, :, :);
 newSlice = reshape(newSlice, [nx, ny]);