C++ (Linux Platform)

Linux SDK supports AArch64, ARMv7hf, RaspberryPi, i686 and x86-64 hardware platforms, and provides corresponding library files for developers to compile programs. Linux SDK does not provide executable applications, the developers can generate executable applications by compiling sample programs.

The following is an example of setting up the development environment for Percipio depth camera on Linux platform, using Ubuntu 20.04:

Download Camport4 SDK
Install Dependencies
Option: Set USB Camera Access Permissions
Compile Sample Programs
Run Sample Programs

Download Camport4 SDK

Choose any of the following methods to download Camport4 SDK:

Access https://github.com/percipioxyz/camport4.git using a browser and download the Camport4 SDK.
Use the git command to obtain a local copy: Open the terminal, switch to the directory where the SDK should be stored, and enter the git clone command to clone the remote repository.
```
git clone https://github.com/percipioxyz/camport4.git
```

The directory structure of Camport4 SDK is shown in the following figure:

Doc Directory

It stores the SDK API reference documentation. The documentation is generated by Doxygen and includes descriptive information about classes, structures and interfaces.

include Directory

It stores the header files used to call dynamic libraries:

TYImageProc.h: the header file for the image post-processing function APIs.
TYCoordinateMapper.h: the header file for mapping of image space transformation.
TY_API.h: the header file for configuring the depth camera and obtaining image data. Users can develop their own applications based on these APIs.
TYVer.h: the header file for retrieving SDK version information.
TYParameter.h: the header file for the GenICam standard APIs.
TYDefs.h: the header file for Camport SDK data types.

lib/linux Directory

It stores library files which support the Linux operating system and are compatible with AArch64, ARMv7hf, RaspberryPi, i686 and x86-64 hardware platforms.

lib/win Directory

It stores library files which support the Windows operating system and are compatible with x64 and x86 hardware platforms.

sample Directory

It contains subfolders including cloud_viewer, common, sample_v1, sample_v2 and sample_genicam_sfnc, where sample_v1, sample_v2 and sample_genicam_sfnc include sample programs provided for users to refer to when programming. For more details about the sample programs, please refer to Sample Program Descriptions.

Install Dependencies

Install Cmake

Install Cmake by inputting the following command in the terminal:
```
sudo apt-get install cmake
```
Check if the installation is successful by checking the version information.
```
cmake --version
```

Note

The recommended CMake version is 3.24. Other versions are not tested and the compatibility is not guaranteed.

Install OpenCV (Optional)

sudo apt-get install libopencv-dev

Note

The recommended OpenCV version is 4.5.5. Other versions are not tested and the compatibility is not guaranteed.
If the target application does not use OpenCV for image display or other image processing calculations, there is no need to install OpenCV.

Install USB Driver (Optional)

When you are using a USB camera, installing the USB driver is necessary for compiling and running the Camport SDK.

sudo apt-get install libusb-1.0-0-dev

Set USB Camera Access Permissions

According to the permission management mechanism of Linux system, by default, root permission is required to access the Percipio USB camera. Non-root users can create a udev rule to modify the device permissions.

The setup steps are as follows:

Direct to directory “etc/udev/rules.d”, create a rule file with the extension .rules (such as 88-tyusb.rules).

Note

The beginning of the rule file name must be a number (0 ~ 99), the larger the number, the higher the priority.
```
sudo nano /etc/udev/rules.d/88-tyusb.rules
```

Add the following rules in the rule file. The example below is to add the USB device to the tofu user group.

SUBSYSTEM== "usb",ATTRS{idProduct}=="1003",ATTRS{idVendor}=="04b4",GROUP="tofu",MODE="0666"
SUBSYSTEM== "usb",ATTRS{idProduct}=="1004",ATTRS{idVendor}=="04b4",GROUP="tofu",MODE="0667"

After restarting the PC, access permission to the USB depth camera will be granted.

Compile Sample Programs

Environment Setup

Enter the camport4 directory and execute the following command to copy the necessary library files and prepare for compilation.
```
sudo cp lib/linux/lib_x64/libty* /usr/lib/
cd sample
mkdir build
cd build
```
Build Execution

Run the following commands to compile. Replace /home/{SDK-path} with your actual SDK installation path.
```
cmake .. -DTYCam_DIR=/home/{YOUR_SDK_PATH}/camport4
make
```
Configuring Build Options for Sample Programs

The commands above compile all sample programs (sample_genicam_sfnc, sample_v1, sample_v2) by default.

You can customize the build by adding the following parameters to the cmake command based on your needs:
- To exclude sample_v1 and sample_genicam_sfnc:
```
cmake .. -DBUILD_SAMPLES=off
```
- To exclude sample_v2:
```
cmake .. -DBUILD_CPLUSPLUS_SAMPLE=off
```
- To exclude sample_genicam_sfnc:
```
cmake .. -DBUILD_SAMPLE_GENICAM_SFNC=off
```
- To compile sample_v2 without OpenCV dependency:
```
cmake .. -DBUILD_SAMPLES=off -DBUILD_SAMPLE_V2_WITH_OPENCV=off
```
Configuring Build Options for Image Display

Image display is enabled by default during the CMake configuration. The build system will select a display backend automatically in the following order of preference:
1. OpenCV is used first if it is detected on the system.
2. If OpenCV is not found, the system falls back to using the lighter-weight, native Linux X11 interface.
To disable image display entirely, add this parameter to your cmake command:
```
cmake .. -DBUILD_SAMPLE_WITH_GUI=OFF
```

The compiled executable files will be generated in the camport4/sample/build/bin directory.

Run Sample Programs

After connecting to the computer, the depth camera will initiate system initialization. Once the power indicator light stabilizes to a steady 1-second on, 1-second off blinking pattern, you can run the following command with root privileges to receive and view real-time depth maps.

sudo ./SimpleView_FetchFrame