tami/MYNT-EYE-S-SDK
6
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

doc(*): remove tools things.

Browse Source
This commit is contained in:
TinyO 2019-08-30 11:23:05 +08:00
parent 8ae150b516
commit 2482ff5bf6
7 changed files with 0 additions and 438 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

1
docs/src/sdk/contents.rst
View File

@ -8,6 +8,5 @@ SDK
sdk_install
data/contents
control/contents
tools/contents
project/contents
sdk_changelog

58
docs/src/sdk/tools/analytics_dataset.rst
View File

@ -1,58 +0,0 @@
.. _analytics_dataset:
Recording Data Sets
====================
The SDK provides the tool ``record`` for recording data sets. Tool details can be seen in `tools/README.md <https://github.com/slightech/MYNT-EYE-S-SDK/tree/master/tools>`_ .
Reference run command:
.. code-block:: bash
./tools/_output/bin/dataset/record2
# Windows
.\tools\_output\bin\dataset\record2.bat
Reference run results on Linux:
.. code-block:: bash
$ ./tools/_output/bin/dataset/record
I0513 21:28:57.128947 11487 utils.cc:26] Detecting MYNT EYE devices
I0513 21:28:57.807116 11487 utils.cc:33] MYNT EYE devices:
I0513 21:28:57.807155 11487 utils.cc:37] index: 0, name: MYNT-EYE-S1000
I0513 21:28:57.807163 11487 utils.cc:43] Only one MYNT EYE device, select index: 0
I0513 21:28:57.808437 11487 channels.cc:114] Option::GAIN: min=0, max=48, def=24, cur=24
I0513 21:28:57.809999 11487 channels.cc:114] Option::BRIGHTNESS: min=0, max=240, def=120, cur=120
I0513 21:28:57.818678 11487 channels.cc:114] Option::CONTRAST: min=0, max=255, def=127, cur=127
I0513 21:28:57.831529 11487 channels.cc:114] Option::FRAME_RATE: min=10, max=60, def=25, cur=25
I0513 21:28:57.848914 11487 channels.cc:114] Option::IMU_FREQUENCY: min=100, max=500, def=200, cur=500
I0513 21:28:57.865185 11487 channels.cc:114] Option::EXPOSURE_MODE: min=0, max=1, def=0, cur=0
I0513 21:28:57.881434 11487 channels.cc:114] Option::MAX_GAIN: min=0, max=48, def=48, cur=48
I0513 21:28:57.897598 11487 channels.cc:114] Option::MAX_EXPOSURE_TIME: min=0, max=240, def=240, cur=240
I0513 21:28:57.913918 11487 channels.cc:114] Option::DESIRED_BRIGHTNESS: min=0, max=255, def=192, cur=192
I0513 21:28:57.930177 11487 channels.cc:114] Option::IR_CONTROL: min=0, max=160, def=0, cur=0
I0513 21:28:57.946341 11487 channels.cc:114] Option::HDR_MODE: min=0, max=1, def=0, cur=0
Saved 1007 imgs, 20040 imus to ./dataset
I0513 21:29:38.608772 11487 record.cc:118] Time beg: 2018-05-13 21:28:58.255395, end: 2018-05-13 21:29:38.578696, cost: 40323.3ms
I0513 21:29:38.608853 11487 record.cc:121] Img count: 1007, fps: 24.9732
I0513 21:29:38.608873 11487 record.cc:123] Imu count: 20040, hz: 496.983
Results save into ``<workdir>/dataset`` by default. You can also add parameter, select other directory to save.
Record contents:
.. code-block:: none
<workdir>/
└─dataset/
├─left/
│ ├─stream.txt # Image infomation
│ ├─000000.png # Image，index 0
│ └─...
├─right/
│ ├─stream.txt # Image information
│ ├─000000.png # Image，index 0
│ └─...
└─motion.txt # IMU information

38
docs/src/sdk/tools/analytics_imu.rst
View File

@ -1,38 +0,0 @@
.. _analytics_imu:
Analyzing IMU
==============
The SDK provides the script ``imu_analytics.py`` for IMU analysis. The tool details can be seen in `tools/README.md <https://github.com/slightech/MYNT-EYE-S-SDK/tree/master/tools>`_ .
Refer to run commands and results on Linux:
.. code-block:: bash
$ python tools/analytics/imu_analytics.py -i dataset -c tools/config/mynteye/mynteye_config.yaml -al=-1.2,1.2 -gl= -gdu=d -gsu=d -kl=
imu analytics ...
input: dataset
outdir: dataset
gyro_limits: None
accel_limits: [(-1.2, 1.2), (-1.2, 1.2), (-1.2, 1.2), (-1.2, 1.2)]
time_unit: None
time_limits: None
auto: False
gyro_show_unit: d
gyro_data_unit: d
temp_limits: None
open dataset ...
imu: 20040, temp: 20040
timebeg: 4.384450, timeend: 44.615550, duration: 40.231100
save figure to:
dataset/imu_analytics.png
imu analytics done
The analysis result graph will be saved in the data set directory, as follows:
.. image:: ../../../images/sdk/tools/imu_analytics.png
In addition, the script specific options can be executed ``-h``:
.. code-block:: bash
$ python tools/analytics/imu_analytics.py -h

55
docs/src/sdk/tools/analytics_stamp.rst
View File

@ -1,55 +0,0 @@
.. _analytics_stamp:
Analyze Time Stamps
====================
SDK provides a script for timestamp analysis ``stamp_analytics.py`` . Tool details are visible in `tools/README.md <https://github.com/slightech/MYNT-EYE-S-SDK/tree/master/tools>`_ .
Reference run commands and results on Linux:
.. code-block:: bash
$ python tools/analytics/stamp_analytics.py -i dataset -c tools/config/mynteye/mynteye_config.yaml
stamp analytics ...
input: dataset
outdir: dataset
open dataset ...
save to binary files ...
binimg: dataset/stamp_analytics_img.bin
binimu: dataset/stamp_analytics_imu.bin
img: 1007, imu: 20040
rate (Hz)
img: 25, imu: 500
sample period (s)
img: 0.04, imu: 0.002
diff count
imgs: 1007, imus: 20040
imgs_t_diff: 1006, imus_t_diff: 20039
diff where (factor=0.1)
imgs where diff > 0.04*1.1 (0)
imgs where diff < 0.04*0.9 (0)
imus where diff > 0.002*1.1 (0)
imus where diff < 0.002*0.9 (0)
image timestamp duplicates: 0
save figure to:
dataset/stamp_analytics.png
stamp analytics done
The analysis result graph will be saved in the dataset directory, as follows:
.. image:: ../../../images/sdk/tools/stamp_analytics.png
In addition, the script specific options can be executed ``-h`` to understand:
.. code-block:: bash
$ python tools/analytics/stamp_analytics.py -h
.. tip::
Suggestions when recording data sets ``record.cc`` annotation display image inside ``cv::imshow()``, ``dataset.cc`` annotation display image inside ``cv::imwrite()`` . Because these operations are time-consuming, they can cause images to be discarded. In other words, consumption can't keep up with production, so some images are discarded. ``GetStreamDatas()`` used in ``record.cc`` only caches the latest 4 images.

9
docs/src/sdk/tools/contents.rst
View File

@ -1,9 +0,0 @@
SDK Tools
==================
.. toctree::
:maxdepth: 2
analytics_dataset
analytics_imu
analytics_stamp

269
docs/src/tools/calibration_tool.rst
View File

@ -1,269 +0,0 @@
.. _calibration_tool:
Calibration Tool Manual
=======================
Introduction
--------
1.1 Support Platform
--------
Currently the calibration tool only supports Ubuntu 16.04 LTS, but support the official, ROS multiple version of OpenCV dependencies.
==================== ==================== ======================
Platform Architecture Different dependence
==================== ==================== ======================
Ubuntu 16.04 LTS x64(amd64) libopencv-dev
Ubuntu 16.04 LTS x64(amd64) ros-kinetic-opencv3
==================== ==================== ======================
1.2 Tools description
--------
Deb/ppa installation package is available on Ubuntu. The architecture, dependencies, and versions will be distinguished from the name:
* mynteye-s-calibrator-opencv-official-1.0.0_amd64.deb
* mynteye-s-calibrator-opencv-ros-kinetic-1.0.0_amd64.deb
======================= ==================== ============================================================
Dependency identifier Dependency package Detailed description
======================= ==================== ============================================================
opencv-official libopencv-dev https://packages.ubuntu.com/xenial/libopencv-dev
opencv-ros-kinetic ros-kinetic-opencv3 http://wiki.ros.org/opencv3
======================= ==================== ============================================================
1.3 Deb Toolkit Get
--------
==================== ========================================================================
Method of Obtaining Get address
==================== ========================================================================
Baidu Cloud https://pan.baidu.com/s/19rW0fPKUlQj6eldZpZFoAA Extraction code: a6ps
Google Drive https://drive.google.com/open?id=1RsV2WEKAsfxbn-Z5nGjk5g3ml1UDEsDc
==================== ========================================================================
Installation
--------
2.1 Installation Preparation
--------
* Ubuntu 16.04 LTS environment, x64 architecture
* Deb package for the calibration tool, select OpenCV dependencies as needed
(this step is not required for PPA installation)
2.2 Install ppa Package
--------
.. code-block:: bash
$ sudo add-apt-repository ppa:slightech/mynt-eye-s-sdk
$ sudo apt-get update
$ sudo apt-get install mynteye-s-calibrator
$ sudo ln -sf /opt/myntai/mynteye-s-calibrator/mynteye-s-calibrator /usr/local/bin/ mynteye-s-calibrator
2.3 Install deb Package
--------
Install the deb package with udo dpkg -i:
.. code-block:: bash
$ sudo dpkg -i mynteye-s-calibrator-opencv-official-1.0.0_amd64.deb
...
(Reading database ... 359020 files and directories currently installed.)
Preparing to unpack mynteye-s-calibrator-opencv-official-1.0.0_amd64.deb ...
Unpacking mynteye-s-calibrator (1.0.0) over (1.0.0) ...
Setting up mynteye-s-calibrator (1.0.0) ...
If you encounter an error that the dependency package is not installed, for example:
.. code-block:: bash
$ sudo dpkg -i mynteye-s-calibrator-opencv-official-1.0.0_amd64.deb
Selecting previously unselected package mynteye-s-calibrator.
(Reading database ... 358987 files and directories currently installed.)
Preparing to unpack mynteye-s-calibrator-opencv-official-1.0.0_amd64.deb ...
Unpacking mynteye-s-calibrator (1.0.0) ...
dpkg: dependency problems prevent configuration of mynteye-s-calibrator:
mynteye-s-calibrator depends on libatlas-base-dev; however:
Package libatlas-base-dev is not installed.
dpkg: error processing package mynteye-s-calibrator (--install):
dependency problems - leaving unconfigured
Errors were encountered while processing:
mynteye-s-calibrator
You can continue use sudo apt-get -f install to finished install
.. code-block:: bash
$ sudo apt-get -f install
Reading package lists... Done
Building dependency tree
Reading state information... Done
Correcting dependencies... Done
The following additional packages will be installed:
libatlas-base-dev
Suggested packages:
libblas-doc liblapack-doc
The following NEW packages will be installed:
libatlas-base-dev
0 upgraded, 1 newly installed, 0 to remove and 0 not upgraded.
1 not fully installed or removed.
Need to get 3,596 kB of archives.
After this operation, 30.8 MB of additional disk space will be used.
Do you want to continue? [Y/n]
Get:1 http://cn.archive.ubuntu.com/ubuntu xenial/universe amd64 libatlas-base-dev amd64 3.10.2-9 [3,596 kB]
Fetched 3,596 kB in 3s (1,013 kB/s)
Selecting previously unselected package libatlas-base-dev.
(Reading database ... 358993 files and directories currently installed.)
Preparing to unpack .../libatlas-base-dev_3.10.2-9_amd64.deb ...
Unpacking libatlas-base-dev (3.10.2-9) ...
Setting up libatlas-base-dev (3.10.2-9) ...
update-alternatives: using /usr/lib/atlas-base/atlas/libblas.so to provide /usr/lib/libblas.so (libblas.so) in auto mode
update-alternatives: using /usr/lib/atlas-base/atlas/liblapack.so to provide /usr/lib/liblapack.so (liblapack.so) in auto mode
Setting up mynteye-s-calibrator (1.0.0) ...
How To Use
--------
3.1 Preparation For Use
--------
* MYNT EYE S Camera
* Checkerboard
* Evenly illuminated scene
3.2 Use Command
--------
* After installing the calibration tool, you can run the `mynteye-s-calibrator` command directly on the terminal to calibrate. -h can see its options:
.. code-block:: bash
$ mynteye-s-calibrator -h
Usage: mynteye-s-calibrator [options]
help: mynteye-s-calibrator -h
calibrate: mynteye-s-calibrator -x 11 -y 7 -s 0.036
Calibrate MYNT EYE S device.
Options:
-h, --help show this help message and exit
-x WIDTH, --width=WIDTH The chessboard width, default: 11
-y HEIGHT, --height=HEIGHT The chessboard height, default: 7
-s METERS, --square=METERS The chessboard square size in meters, default: 0.036
-n NUMBER, --number=NUMBER The number of imagestools to use for calibration, default: 11
-p PATH, --path=PATH The path to save the result, default: folder name using device's SN
* -x -y -s Used to set the width, height, and grid size of the calibration plate. Width and height refer to the number of black and white intersections in the horizontal and vertical directions of the checkerboard. Square size in meters.
3.3 Steps For Usage
--------
* First, connect the MYNT EYE S camera.
* Then, run the mynteye-s-calibrator <calibration board parameter> command in the terminal.
.. image:: ../../images/tools/calibration001.png
:width: 60%
* Follow the prompts to select an index for the camera's resolution, perform image calibration at this resolution
* The S1030 camera only need calibrate 752*480 resolution. The S2100 camera need calibrate 2560*800 and 1280*400 resolutions.
* As far as possible, let the calibration plate cover the left and right eye images of the camera,
and take care of the surroundings (maximum distortion). The calibration tool will automatically
evaluate the qualified image for the calibration calculation and will indicate on the terminal how many have been selected.
Reference acquisition image, as follows:
.. image:: ../../images/tools/calibration002.png
:width: 60%
.. image:: ../../images/tools/calibration003.png
:width: 60%
.. image:: ../../images/tools/calibration004.png
:width: 60%
.. image:: ../../images/tools/calibration005.png
:width: 60%
.. image:: ../../images/tools/calibration006.png
:width: 60%
* Note: p_x, p_y, size, skew respectively indicate the scale of the calibration plate on the x-axis, y-axis,
zoom, and tilt when the image is acquired. Make a point for reference.
* Once the number of images acquired by the calibration needs is reached,
the calibration calculation will be performed. The output is as follows:
.. image:: ../../images/tools/calibration007.png
:width: 60%
* 1. The terminal will print out the left and right purpose calibration results.
* 2. The calibration results will be written into the files in <SN number> directory.
a) camera_left.yaml: Left eye parameter
b) camera_right.yaml: Right eye parameter
c) extrinsics.yaml: Binocular external parameter
d) img.params.equidistant: Camera parameters, which can be used for S SDK writing
e) stereo_reprojection_error.yaml: Reprojection error
* Finally, you will also be asked if you want to write to the camera device. Enter or `y` to confirm
.. image:: ../../images/tools/calibration008.png
:width: 60%
* After writing to the device, you will be prompted with "Write to device done".
3.4 Calibration result
--------
Calibration result, It is desirable to have a reprojection error of 0.2 or less. If exceeds 1, it needs to be recalibrated.
Reprojection error, visible output after calibration completion "Final reprojection error: 0.201
Pixels", or see the calibration result file "stereo_reprojection_error.yaml".

8
docs/src/tools/contents.rst
View File

@ -1,8 +0,0 @@
.. _tools:
TOOLS SUPPORT
==============
.. toctree::
calibration_tool