[OpenCV] 車道辨識

這兩篇蠻有幫助的，但看起來是同一篇QQ
* https://medium.com/@mrhwick/simple-lane-detection-with-opencv-bfeb6ae54ec0
* https://campushippo.com/lessons/detect-highway-lane-lines-with-opencv-and-python-21438a3e2

[RPi] MLX90640

https://makersportal.com/blog/2020/6/8/high-resolution-thermal-camera-with-raspberry-pi-and-mlx90640

[RPi] MPU9250

https://makersportal.com/blog/2019/11/11/raspberry-pi-python-accelerometer-gyroscope-magnetometer

這篇只能用 python2，不能用 python3
https://maker.pro/raspberry-pi/tutorial/how-to-interface-an-imu-sensor-with-a-raspberry-pi

這篇加減看
https://raspberrypi.stackexchange.com/questions/108896/what-is-rpis-i2c-maximum-speed

[RPi] Object Detection on Pi

這篇可以用欸
https://iotdesignpro.com/projects/raspberry-pi-object-detection-using-tensorflow-and-opencv

[RPi] WiFi Direct on Raspberry Pi

這篇討論看起來是可以用的？
https://www.raspberrypi.org/forums/viewtopic.php?f=63&t=31596&start=25#p1224705

[RPi] Direct Capture Mode on Linux

# pi
sudo apt-get install realvnc-vnc-server
sudo raspi-config
> 5. Interfacing Options.
> VNC
> Yes

# Ubuntu
如果直接用 vncviewer x.x.x.x 會報錯
No configured security type is supported by 3.3 VNC Viewer

要安裝 Standalone 版本，下載完 chmod 成執行檔以後使用
https://www.realvnc.com/en/connect/download/viewer/linux/
下載檔案：VNC-Viewer-6.20.529-Linux-x6.deb
執行檔：vnc-viewer-6.20.529-Linux-x64

https://help.realvnc.com/hc/en-us/articles/360002249917-VNC-Connect-and-Raspberry-Pi

[RPi] Pi4上的晶片

這篇有時候會用到
https://www.electronics-lab.com/project/raspberry-pi-4-look-hood-make/

[BLE] 相關的 Python 套件

iBeacon-Scanner 裡面的 blescan.py 是從這邊來的
https://github.com/adamf/BLE

而該作者久沒更新，建議參考 bleak。不過這套件大量使用 asyncio，沒先講過又不好...

[Linux] 在Linux顯示Android手機螢幕

之前是用 seven-square，現在改用 Vysor，是 chromium 的 plugin

[IoT] RS485

找時間測試一下吧 QQ
* http://www.sonbest.com/sonbestproducts/instrument/record/sr/SD5110B.html
* http://www.sonbest.com/uploads/soft/150309/2-150309133532.pdf
* http://www.sonbest.com/uploads/soft/150309/2-150309115137.pdf
* http://www.sonbus.com/upload/Pdf_cn/SD5110B.pdf

[RPI] HQ Camera

6mm 的要有點天份才能對焦

* https://www.raspberrypi.org/forums/viewtopic.php?f=43&t=272658&start=50#p1656092
* https://www.raspberrypi.org/forums/viewtopic.php?t=272917

[RPI] 8G

Pi 4 出了 8G 版本，終於可以在上面直接使用 TensorFlow 訓練。

如果跑 Donkey Car 專案，總拍照數目約 27xx 張，訓練 Linear  模型，約 85 分跑完。

real    85m53.544s
user    246m6.465s
sys    39m45.854s

但是，如果用 Google Colab 大概五分鐘就可以跑完 QQ