added 70 characters in body
Source Link
xslittlegrass
  • 27.1k
  • 9
  • 94
  • 184

The neural network contains two parts, the feature extraction network and the classification network. For the feature extraction network, we use a simple CNN assemblesresembles the VGG network.

classes = Join[ToString /@ Range[0, 9], ToUpperCase@Alphabet[]];
net = NetGraph[{convnet, 36, SoftmaxLayer[], 36, SoftmaxLayer[], 36, 
   SoftmaxLayer[], 36, 
   SoftmaxLayer[]}, {NetPort["Image"] -> 
    1 -> 2 -> 3 -> NetPort["digit1"], 
   1 -> 4 -> 5 -> NetPort["digit2"], 1 -> 6 -> 7 -> NetPort["digit3"],
    1 -> 8 -> 9 -> NetPort["digit4"]}, 
  "digit1" -> NetDecoder[{"Class", classes}], 
  "digit2" -> NetDecoder[{"Class", classes}], 
  "digit3" -> NetDecoder[{"Class", classes}], 
  "digit4" -> NetDecoder[{"Class", classes}]]

The neural network contains two parts, the feature extraction network and the classification network. For the feature extraction network, we use a simple CNN assembles the VGG network.

net = NetGraph[{convnet, 36, SoftmaxLayer[], 36, SoftmaxLayer[], 36, 
   SoftmaxLayer[], 36, 
   SoftmaxLayer[]}, {NetPort["Image"] -> 
    1 -> 2 -> 3 -> NetPort["digit1"], 
   1 -> 4 -> 5 -> NetPort["digit2"], 1 -> 6 -> 7 -> NetPort["digit3"],
    1 -> 8 -> 9 -> NetPort["digit4"]}, 
  "digit1" -> NetDecoder[{"Class", classes}], 
  "digit2" -> NetDecoder[{"Class", classes}], 
  "digit3" -> NetDecoder[{"Class", classes}], 
  "digit4" -> NetDecoder[{"Class", classes}]]

The neural network contains two parts, the feature extraction network and the classification network. For the feature extraction network, we use a simple CNN resembles the VGG network.

classes = Join[ToString /@ Range[0, 9], ToUpperCase@Alphabet[]];
net = NetGraph[{convnet, 36, SoftmaxLayer[], 36, SoftmaxLayer[], 36, 
   SoftmaxLayer[], 36, 
   SoftmaxLayer[]}, {NetPort["Image"] -> 
    1 -> 2 -> 3 -> NetPort["digit1"], 
   1 -> 4 -> 5 -> NetPort["digit2"], 1 -> 6 -> 7 -> NetPort["digit3"],
    1 -> 8 -> 9 -> NetPort["digit4"]}, 
  "digit1" -> NetDecoder[{"Class", classes}], 
  "digit2" -> NetDecoder[{"Class", classes}], 
  "digit3" -> NetDecoder[{"Class", classes}], 
  "digit4" -> NetDecoder[{"Class", classes}]]
added 193 characters in body
Source Link
xslittlegrass
  • 27.1k
  • 9
  • 94
  • 184

Here is one example using a convolutional neural network (CNN) to crack the CAPTCHA. We will use a CAPTCHA library to generate sample CAPTCHA images and then train a neural network to decode these images. The trained network can be downloaded from crack_CAPTCHA.wlnetcrack_CAPTCHA.wlnet.

A few more sample images can be downloaded from here.

The trained network can be downloaded from herehere

Here is one example using a convolutional neural network (CNN) to crack the CAPTCHA. We will use a CAPTCHA library to generate sample CAPTCHA images and then train a neural network to decode these images. The trained network can be downloaded from crack_CAPTCHA.wlnet.

The trained network can be downloaded from here

Here is one example using a convolutional neural network (CNN) to crack the CAPTCHA. We will use a CAPTCHA library to generate sample CAPTCHA images and then train a neural network to decode these images. The trained network can be downloaded from crack_CAPTCHA.wlnet.

A few more sample images can be downloaded from here.

The trained network can be downloaded from here

added 149 characters in body
Source Link
xslittlegrass
  • 27.1k
  • 9
  • 94
  • 184

Here is one example using a convolutional neural network (CNN) to crack the CAPTCHA. We will use a CAPTCHA library to generate sample CAPTCHA images and then train a neural network to decode these images. The trained network can be downloaded from crack_CAPTCHA.wlnet.

In the following, we will describe how to prepare the training data, construct and train the CNN.

We use the default setting for the hyper-parameters and train 3 rounds. The total training time is about 2.5 hours on the Nvidia 750M in use. Since Mathematica 11.1 doesn't support GPU training on Mac, the network is trained in Mathematica version 11.0.

The trained network can be downloaded from here

Here is one example using a convolutional neural network (CNN) to crack the CAPTCHA. We will use a CAPTCHA library to generate sample CAPTCHA images and then train a neural network to decode these images.

We use the default setting for the hyper-parameters and train 3 rounds. The total training time is about 2.5 hours on the Nvidia 750M in use.

Here is one example using a convolutional neural network (CNN) to crack the CAPTCHA. We will use a CAPTCHA library to generate sample CAPTCHA images and then train a neural network to decode these images. The trained network can be downloaded from crack_CAPTCHA.wlnet.

In the following, we will describe how to prepare the training data, construct and train the CNN.

We use the default setting for the hyper-parameters and train 3 rounds. The total training time is about 2.5 hours on the Nvidia 750M in use. Since Mathematica 11.1 doesn't support GPU training on Mac, the network is trained in Mathematica version 11.0.

The trained network can be downloaded from here

added 1161 characters in body
Source Link
xslittlegrass
  • 27.1k
  • 9
  • 94
  • 184
Loading
added 1161 characters in body
Source Link
xslittlegrass
  • 27.1k
  • 9
  • 94
  • 184
Loading
Source Link
xslittlegrass
  • 27.1k
  • 9
  • 94
  • 184
Loading