如何进行OpenNI-Linux-Arm64-2.3.0.66安装与调试，很多新手对此不是很清楚，为了帮助大家解决这个难题，下面小编将为大家详细讲解，有这方面需求的人可以来学习下，希望你能有所收获。
最近在研究某摄像头，通过通用串行总线插入到Arm64的板子上进行安装调试，这里尝试着把过程记录下来。
1 下载SDK
从官网上下载相应系统和版本的SDK，我这里是Linux-ARM64版本的，下载的是OpenNI-Linux-Arm64-2.3.0.66。或者wget链接地址
proxychains 4 gwet https://abzg-OSS。OSS-中国深圳。阿里云克斯。com/files/OpenNI-Linux-Arm 64-2。3 .0 .66 .活力
[proxy chains]配置文件找到：/etc/proxychains。conf
[proxychains]预加载/usr/local/lib/libproxychains 4。因此
[proychains]DLline : proychains-ng 4.14
-2021-02-0820:57:26-https://abzg-OSS。OSS-中国深圳。阿里云克斯。com/files/OpenNI-Linux-Arm64-2。3 .0 .66 .活力
正在解析主机abzg-OSS。OSS-中国深圳。阿里云克斯。com(abzg-OSS。OSS-中国深圳。阿里云克斯。com).224.0.0.1
正在连接abzg-OSS。OSS-中国深圳。阿里云克斯。com(abzg-OSS。OSS-中国深圳。阿里云克斯。com)| 224。0 .0 .1 | :443.[proxychains]
严格链.172.18.6.135:8001.abzg-OSS。OSS-中国深圳。阿里云克斯。com 36360。36666666666.好
已连接。
已发出超文本传送协议请求，正在等待回应.200OK
长度：3758034(3.6M)[应用程序/zip]
正在保存至：英寸OpenNi-Linux-Arm 64-2。3 .0 .66 .zip "
OpenNI-Linux-arm 64-2。3 .0 .66 .拉链100%
[===============================================]3.58m 7.57 MB/sin 0.5s
2021-02-0820:57:27(7.57兆字节/秒)-已保存“OpenNI-Linux-Arm64-2。3 .0 .66 .zip"[3758034/3758034])解压，目录结构如下
-rw-rw-r-1工作9753 9月2916336054变更。文本文件（textfile）
-rw-rw-r-1工作工作许可
通知
-rw-rw-r-1工作时间5615年9月29日16:54日
1月1215336052日
-rw-rw-r-1工作工作1863年9月2916:54日重新提交。文本文件（textfile）
drwxrwxr-x12workwork4096J
an 12 15:52 Samples
drwxrwxr-x 3 work work 4096 Jan 12 15:52 Tools
drwxrwxr-x 8 work work 4096 Jan 12 15:52 include
-rw-rw-r-- 1 work work 2138 Mar 10 2020 install.sh
-rw-rw-r-- 1 work work 1221 Sep 29 16:54 orbbec-usb.rules
阅读README文件，其中
0 OpenNI-Linux-Arm64-2.3.0.66 > cat README OpenNI ------ Website: http://structure.io/openni Building Prerequisites ====================== ... Linux ----- - GCC 4.x From: http://gcc.gnu.org/releases.html Or via apt: sudo apt-get install g++ - Python 2.6+/3.x From: http://www.python.org/download/ Or via apt: sudo apt-get install python - LibUSB 1.0.x From: http://sourceforge.net/projects/libusb/files/libusb-1.0/ Or via apt: sudo apt-get install libusb-1.0-0-dev - LibUDEV sudo apt-get install libudev-dev - JDK 6.0 From: http://www.oracle.com/technetwork/java/javase/downloads/jdk-6u32-downloads-1594644.html Or via apt: sudo apt-get install openjdk-6-jdk - FreeGLUT3 From: http://freeglut.sourceforge.net/index.php#download Or via apt: sudo apt-get install freeglut3-dev - Doxygen From: http://www.stack.nl/~dimitri/doxygen/download.html#latestsrc Or via apt: sudo apt-get install doxygen - GraphViz From: http://www.graphviz.org/Download_linux_ubuntu.php Or via apt: sudo apt-get install graphviz ... Building ======== Building on Windows: Open the solution OpenNI.sln Building on Linux / OS X: Run: $ make Cross-Compiling for ARM on Linux: The following environment variables should be defined: - ARM_CXX=<path to cross-compilation g++> - ARM_STAGING=<path to cross-compilation staging dir> Then, run: $ PLATFORM=Arm make Building for Android: - Create a workspace in eclipse - Import all projects under root directory Creating OpenNI2 Package ======================== - Go into the directory 'Packaging' - Run ReleaseVersion.py [x86|x64|Arm|Android] - Installer will be placed in the 'Final' directory Building in an armhf qemu-enabled docker container ================================================== - Note: we're building from the osrf fork which has a couple of minor tweaks compared to the orbbec upstream repo (https://github.com/orbbec/OpenNI2/compare/orbbec-dev...osrf:orbbec-dev?expand=1): - don't force soft-float in third-party software (because we're building with hard-float) - disable documentation generation (because we don't need it, and also because doxygen seems to just hang in this environment). - Check out the code (because git doesn't work inside qemu): $ git clone https://github.com/osrf/OpenNI2 $ cd OpenNI2 $ git checkout orbbec-dev $ cd .. - Start docker, mounting in the git repo: $ docker run -ti -v `pwd`/OpenNI2:/tmp/OpenNI2 osrf/ubuntu_armhf:trusty bash - Now inside docker, install prereqs (we're removing udev because it somehow conflicts with libudev-dev; we're leaving out doxygen and graphviz because we're not going to build docs): # apt-get update # apt-get remove -y --force-yes udev # apt-get install -y build-essential python libusb-1.0-0-dev libudev-dev openjdk-6-jdk freeglut3-dev - Still inside docker, build the release package (note that, instead of running `ReleaseVersion.py` inside `Packaging`, you could just run `make HAS_JAVA=1 release` at the top level): # cd /tmp/OpenNI2/Packaging # ./ReleaseVersion.py Arm - If all goes well, there will be a file in the docker container called something like `/tmp/OpenNI2/Packaging/Final/OpenNI-Linux-Arm-2.3.tar.bz2`. It's also in the corresponding place in the host filesystem, so it will survive exiting the container. That file contains the compiled libraries (importantly, `OpenNI-Linux-Arm-2.3/Redist/libOpenNI2.so`).
这里指明了一些软件运行所需的依赖和安装方法，这里就不再赘述。后续的 building 等不是必须的操作。
2 配置
在解压后的目录中，执行 install.sh 文件
sudo sh install.sh
该脚本一共做了两件事，为摄像头设备增加UDEV规则，把开发环境写到 OpenNIDevEnvironment文件中。
cat OpenNIDevEnvironment export OPENNI2_INCLUDE=/home/work/abc/OpenNI-Linux-Arm64-2.3.0.65/Include export OPENNI2_REDIST=/home/work/abc/OpenNI-Linux-Arm64-2.3.0.65/Redist
注意第一个环境变量的Include，实际的目录名称为 include，记得改过来（要么把目录名称改成大写，要么把环境变量改成小写，我这里是选择改环境变量。）
然后再进行加载使其生效。
source OpenNIDevEnvironment
检查一下
env | grep OPENNI2_INCLUDE OPENNI2_INCLUDE=/home/work/abc/OpenNI-Linux-Arm64-2.3.0.66/include
3 测试
在Samples目录下，我们可以看到很多样例
drwxrwxr-x 3 work work 4096 Jan 20 22:10 Bin drwxrwxr-x 2 work work 4096 Jan 20 22:10 ClosestPointViewer drwxrwxr-x 2 work work 4096 Jan 20 22:10 EventBasedRead drwxrwxr-x 2 work work 4096 Jan 20 22:10 MWClosestPoint drwxrwxr-x 2 work work 4096 Jan 20 22:10 MWClosestPointApp drwxrwxr-x 2 work work 4096 Jan 20 22:10 MultiDepthViewer drwxrwxr-x 3 work work 4096 Feb 8 17:28 MultipleStreamRead drwxrwxr-x 3 work work 4096 Jan 21 20:06 SimpleRead drwxrwxr-x 3 work work 4096 Jan 27 10:59 SimpleViewer drwxrwxr-x 3 work work 4096 Jan 20 22:10 SimpleViewer.java
我们就以SimpleRead为例，来测试一下。进入目录，结构如下：
总用量 52 drwxrwxr-x 2 work work 4096 Jan 12 15:52 . drwxrwxr-x 12 work work 4096 Jan 12 15:52 .. -rw-rw-r-- 1 work work 5080 Mar 10 2020 CommonCppMakefile -rw-rw-r-- 1 work work 1710 Mar 10 2020 CommonDefs.mak -rw-rw-r-- 1 work work 538 Mar 10 2020 CommonTargets.mak -rw-rw-r-- 1 work work 631 Mar 10 2020 Makefile -rw-rw-r-- 1 work work 3535 Mar 10 2020 OniSampleUtilities.h -rw-rw-r-- 1 work work 493 Mar 10 2020 Platform.Arm -rw-rw-r-- 1 work work 521 Mar 10 2020 Platform.Arm64 -rw-rw-r-- 1 work work 150 Mar 10 2020 Platform.x64 -rw-rw-r-- 1 work work 531 Mar 10 2020 Platform.x86 -rw-rw-r-- 1 work work 3500 Mar 10 2020 main.cpp
3.1 Makefile
include CommonDefs.mak BIN_DIR = Bin INC_DIRS = \ ../../Include \ ../Common SRC_FILES = *.cpp USED_LIBS += OpenNI2 EXE_NAME = SimpleRead CFLAGS += -Wall ifndef OPENNI2_INCLUDE $(error OPENNI2_INCLUDE is not defined. Please define it or 'source' the OpenNIDevEnvironment file from the installation) else ifndef OPENNI2_REDIST $(error OPENNI2_REDIST is not defined. Please define it or 'source' the OpenNIDevEnvironment file from the installation) endif INC_DIRS += $(OPENNI2_INCLUDE) include CommonCppMakefile .PHONY: copy-redist copy-redist: cp -R $(OPENNI2_REDIST)/* $(OUT_DIR) $(OUTPUT_FILE): copy-redist
把 INC_DIRS 中的 ../../Include \ 改成实际的 ../../include \，而 ../Common 目录其实并不存在，也可以注释掉。
3.2 源码
main.cpp 文件
/*****************************************************************************
* *
* OpenNI 2.x Alpha *
* Copyright (C) 2012 PrimeSense Ltd. *
* *
* This file is part of OpenNI. *
* *
* Licensed under the Apache License, Version 2.0 (the "License"); *
* you may not use this file except in compliance with the License. *
* You may obtain a copy of the License at *
* *
* http://www.apache.org/licenses/LICENSE-2.0 *
* *
* Unless required by applicable law or agreed to in writing, software *
* distributed under the License is distributed on an "AS IS" BASIS, *
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. *
* See the License for the specific language governing permissions and *
* limitations under the License. *
* *
*****************************************************************************/
#include <stdio.h>
#include <OpenNI.h>
#include "OniSampleUtilities.h"
#define SAMPLE_READ_WAIT_TIMEOUT 2000 //2000ms
using namespace openni;
int main(int argc, char* argv[])
{
Status rc = OpenNI::initialize();
if (rc != STATUS_OK)
{
printf("Initialize failed\n%s\n", OpenNI::getExtendedError());
return 1;
}
Device device;
if (argc < 2)
rc = device.open(ANY_DEVICE);
else
rc = device.open(argv[1]);
if (rc != STATUS_OK)
{
printf("Couldn't open device\n%s\n", OpenNI::getExtendedError());
return 2;
}
VideoStream depth;
if (device.getSensorInfo(SENSOR_DEPTH) != NULL)
{
rc = depth.create(device, SENSOR_DEPTH);
if (rc != STATUS_OK)
{
printf("Couldn't create depth stream\n%s\n", OpenNI::getExtendedError());
return 3;
}
}
rc = depth.start();
if (rc != STATUS_OK)
{
printf("Couldn't start the depth stream\n%s\n", OpenNI::getExtendedError());
return 4;
}
VideoFrameRef frame;
while (!wasKeyboardHit())
{
int changedStreamDummy;
VideoStream* pStream = &depth;
rc = OpenNI::waitForAnyStream(&pStream, 1, &changedStreamDummy, SAMPLE_READ_WAIT_TIMEOUT);
if (rc != STATUS_OK)
{
printf("Wait failed! (timeout is %d ms)\n%s\n", SAMPLE_READ_WAIT_TIMEOUT, OpenNI::getExtendedError());
continue;
}
rc = depth.readFrame(&frame);
if (rc != STATUS_OK)
{
printf("Read failed!\n%s\n", OpenNI::getExtendedError());
continue;
}
if (frame.getVideoMode().getPixelFormat() != PIXEL_FORMAT_DEPTH_1_MM && frame.getVideoMode().getPixelFormat() != PIXEL_FORMAT_DEPTH_100_UM)
{
printf("Unexpected frame format\n");
continue;
}
DepthPixel* pDepth = (DepthPixel*)frame.getData();
int middleIndex = (frame.getHeight()+1)*frame.getWidth()/2;
printf("[%08llu] %8d\n", (long long)frame.getTimestamp(), pDepth[middleIndex]);
}
depth.stop();
depth.destroy();
device.close();
OpenNI::shutdown();
return 0;
}
其实是将镜头中心的点所获取到的深度信息不断地进行输出。
3.3 编译
make
输出
mkdir -p Bin/Intermediate/Arm64-Release/SimpleRead g++ -MD -MP -MT "./Bin/Intermediate/Arm64-Release/SimpleRead/main.d Bin/Intermediate/Arm64-Release/SimpleRead/main.o" -c -march=armv8-a -O3 -fno-tree-pre -fno-strict-aliasing -ftree-vectorize -ffast-math -funsafe-math-optimizations -flax-vector-conversions -Wall -O2 -DNDEBUG -I../../include -I/home/work/abc/OpenNI-Linux-Arm64-2.3.0.66/include -DXN_NEON -fPIC -fvisibility=hidden -Werror -o Bin/Intermediate/Arm64-Release/SimpleRead/main.o main.cpp mkdir -p Bin/Arm64-Release cp -R /home/work/abc/OpenNI-Linux-Arm64-2.3.0.66/Redist/* Bin/Arm64-Release g++ -o Bin/Arm64-Release/SimpleRead ./Bin/Intermediate/Arm64-Release/SimpleRead/main.o -LBin/Arm64-Release -lOpenNI2 -Wl,-rpath ./
会发现目录下多了一个 Bin 目录
├── Bin │ ├── Arm64-Release │ │ ├── OpenNI.ini │ │ ├── OpenNI2 │ │ ├── SimpleRead │ │ ├── libDepthUtils.a │ │ ├── libOpenNI2.so │ │ └── libSonixCamera.a │ └── Intermediate │ └── Arm64-Release ├── CommonCppMakefile ├── CommonDefs.mak ├── CommonTargets.mak ├── Makefile ├── OniSampleUtilities.h ├── Platform.Arm ├── Platform.Arm64 ├── Platform.x64 ├── Platform.x86 └── main.cpp
3.4 执行
0 SimpleRead > ./Bin/Arm64-Release/SimpleRead Warning: USB events thread - failed to set priority. This might cause loss of data... [00000000] 3501 [00033688] 3501 [00067376] 3537 [00101065] 3501 [00303195] 3501 [00336884] 3537 [00370572] 3501 [00404261] 3501 . . .
可以看到，这里就输出了深度信息。
至此，我们就把OpenNI在Linux/Ubuntu16.04-Arm64上安装成功了。
看完上述内容是否对您有帮助呢？如果还想对相关知识有进一步的了解或阅读更多相关文章，请关注行业资讯频道，感谢您对的支持。
内容来源网络，如有侵权，联系删除，本文地址：https://www.230890.com/zhan/41716.html