In the remainder of this blog post I’ll explain what the Intersection over Union evaluation metric is and why we use it.
I’ll also provide a Python implementation of Intersection over Union that you can use when evaluating your own custom object detectors.
Finally, we’ll look at some actual results of applying the Intersection over Union evaluation metric to a set of ground-truth and predicted bounding boxes.
Build meta data by lableImage tool
Convert to json fomat
[
{
"formatid":1,
"width" : 1024,
"heigh": 800,
"Blockes": [
{
"x": 50,
"y": 50,
"width" : 50,
"heigh" : 100
}
]
}
]
pip install -r requirements.txt
python detecIoU.py
conda install -c conda-forge tesserocr
python tesseract2wordLevel.py