I think we should add a paper which is named "Detecting Text in Natural Scenes with
Stroke Width Transform",although,it is very early,lots of classical papers mention it.
address:https://www.microsoft.com/en-us/research/wp-content/uploads/2016/02/201020CVPR20TextDetection.pdf

I want to integrate the model programmatically, how to receive more detailed information from the output model. How are the tensors named and structured in the graph to receive information like this (if available):

input:

image data bytes

output:

bounding box location
score/probability
single character location

./build/examples/cpp_recognition/recognition.bin data/captcha/1.png examples/crnn/deploy.prototxt examples/crnn//model/crnn_captcha.caffemodel
F0222 11:25:31.748042 4965 net.cpp:759] Cannot copy param 0 weights from layer 'conv0'; shape mismatch. Source param shape is 64 3 3 3 (1728); target param shape is 64 1 3 3 (576). To learn this layer's parameters from scratch rather than copying fro

after the txt file is saved, pop the the location of the saved txt file in the pop up that appears.