Opencv Installation

I am getting this error in homeassistant:

No OpenCV library found! Install or compile for your system following instructions here:

But as far as I know I have it already installed. What am I missing?

I am not really sure what to do next. Anyone has an idea?

Recommend pip install opencv-python

Please provide more detail? What are you running home assistant on? How did you install opencv? Where are you seeing that error and what are you doing that produces it?

In most Linux distributions, a library is split in to two separate packages. The main package provides the library itself. A separate “dev” package provides the source code files needed to compile code that uses the library. If home assistant is unable to build the python module for opencv it’s incredibly likely that’s because the library was installed but the header files in the dev package can’t be found. But we won’t know for sure without a detailed description of your issue.

Oh, I have this fixed. I have installed opencv as @robmarkcole suggested and I think that was the problem. Not quite sure since then. But I am ok with the component now. Thank you.

1 Like

I have experience with deepstack and I would like to try out OpenCV, but I was faced with the fact that there is no sensible instruction on OpenCV and with examples of how to train OpenCV for face recognition and how to assign a person’s name to this person. For example, you can upload an image from deepstack and specify the name of the person, and the name of the person will be displayed in the Home Assistant.

What I’m interested in about OpenCV:

  1. How do I configure OpenCV in Home Assistant for face recognition?
  2. How to output an event in the Home Assistant as in deepstack?
  3. How to train the system so that the person’s name can be determined by the face?
  4. How and where should this be specified so that the system determines the name of the person, as in the case of deepstack?

Я имею опыт с deepstack и мне бы хотелось опробовать OpenCV, но столкнулся с тем, что нет толковой инструкции по OpenCV и с примерами как обучить OpenCV для распознавания лиц и как присвоить этому лицу имя человека. К примеру у deepstack можно загрузить изображение и указать имя лица, а в Home Assistant будет прилетать имя человека.

Что меня интересует про OpenCV:

  1. Как настроить OpenCV в Home Assistant для распознавания лиц?
  2. Как выводить событие в Home Assistant как у deepstack?
  3. Как обучить систему, чтобы по лицу можно было определить имя человека?
  4. Как и где это нужно указывать, чтобы система определяла имя лица, как в случае deepstack?

I am wondering the exact same thing as you. Did you ever got an answer to this?


No, I never got an answer. I use deepstack