aboutsummaryrefslogtreecommitdiffstats
path: root/README.md
blob: 6e6fe08cecc079b0b344651c2192fa492fd849f4 (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
# Why O-3000
O-3000 is our answer to rigid and inflexible vision solutions. Maybe you know that. The camera you'd like to use
does not support your (embedded) operating system. The interface specification is top secret and certainly you won't
get any driver source code. Due to such experiences we developed the O-3000 cameras. In the meantime, O-3000
has grown into a comprehensive vision system. Everything is open-source and all interface specifications are
freely available. These cameras are developed and produced in Switzerland.

## Color Image Processing Pipeline
The pipeline applies serveral algorithms to captured images from the O-3000 camera. Following pipeline
stages are implemented:
<ol>
<li>Debayering</li>
<li>Auto white balancing</li>
<li>Color Correction</li>
<li>Lense correction</li>
<li>Sharpening</li>
<li>Gamma correction</li>
</ol>


### Build and install
```
git clone https://stettbacher.ch/gitlab/o-3000/driver.git
cd color-pipe
mkdir build
cd build
cmake ..
sudo make install
```


### Cross-Compile
A toolchain file is needed to specify the cross-compiler environment. As a starting point,
use the template 'cmake/toolchain_file_template.cmake'. Specify the cross-compiler and path. 
Run cmake within build directory as follow:

```
cmake -DCMAKE_TOOLCHAIN_FILE=../cmake/toolchain_file_template.cmake ..
```