Ideal images for Viola-Jones Haar cascade in OpenCV
I am trying to detect specific objects in images using Haar cascade in OpenCV.
Let's say I am interested in detecting stop signs in landscape images. When defining positive image samples for my training set, which would be the best kind of image: (a) full images with my object, (b) a medium crop or (c) a tight crop?
Similarly, what's best for negative images? Does this influence overfitting? I would also appreciate any other general tips from those with experience. Thanks.
Image ref: http://kaitou-ace.deviantart.com/art/Stop-sign-on-a-country-road-Michigan-271990933
You only want features that you want to detect in your positive samples. So the C image would be correct for positive samples.
As for negative samples you want EVERYTHING else. Although that is obviously unrealistic if you are using your detector in a specific environment then training to detect that as negative is the right way to go. I.e. lots of pictures of landscapes etc (ones that don't have stop signs in)
- How to pick just one item from a generator?
- Python: Get unbound class method
- global frame vs. stack frame
- How to generate a snapshot of a field in a time step with VTK and Python
- How to read the first letter from the last line in a txt file in python
- How to control scientific notation in matplotlib?
- Streamlit multiselect, if I don't select anything, doesn't show data frame
- How to extend a class in python?
- Is there a standard location to store function cache files in Python?
- C++ function (Vectors) wrapped with Cython being around 4 times slower than equivalent Cython function (NumPy Arrays MemoryViews), with large arrays
- Error in anyjson setup command: use_2to3 is invalid
- Send paid media aiogram 3.10
- Is there a workaround for adding Microsoft Word footnotes dynamically in Python?
- Training a Keras model to identify leap years
- Overload a method based on init variables
- How do I create a constant in Python?
- What is gettext_lazy on django for?
- Pydantic - parse a list of objects from YAML configuration file
- How to print stdout excerpt in IPython
- What is the difference between Spyder and Jupyter?
- How do I create a multiline plot using seaborn?
- How to read the request body using orjson library in FastAPI?
- Does iPython have built-in support for viewing a variable in pager?
- cropping the image by removing the white spaces
- Verbose level with argparse and multiple -v options
- How to return data in JSON format using FastAPI?
- Rounding a rational number to the nearest integer, with half-up
- Python inspector ignores property return hint when using TypeVar
- How to highlight values per column in Polars
- Create arbitrary multidimensional zeros array