Chessmen Position Recognition Using Artificial Neural Networks

1
Chessmen Position Recognition Using Artificial
Neural Networks

• Jun Hou
• Dec. 8, 2003.

2
Scenario

• Augmented Reality chess game detect the
  position of all black chess pieces

• Problem moving camera, hidden pieces
• Constraints 1-2 piece moves each time
• Initial position known
• Problem reduced to whether there is a piece on a
  given square

Fig 1. Illustration of hidden pieces

• Task Generate synthesized images, train ANN,
  test recognition rate

3
Feature Selection

• Normalized chessboard squares
• In 2D find region of interest the chessboard,
  calculate 3D positions.
• In 3D divide each chessboard square into mm
  smaller squares (64mm)
• Map 3D square positions onto 2D
• Use average of each square as input

• Camera angles and chessboard square positions
• To compensate for the black ratio difference

• Prior probability of each square occupied by a
  chess piece

4
Neural Network Design

• ANN applied
• Tradition GD
• Gradient Descent with variable learning rate
• Gradient Descent with momentum

Fig 2. Neural Network Design
5
Experimental Results Data generation and
preprocessing

• Use Blender Version 2.30 to model the pieces and
  use Python scripts to generate the synthesized
  images
• Change the chessboard state play one random
  move each time, and observe the chessboard from
  different viewpoints
• Image properties clean, high contrast, little
  noise, no distortion, no lighting variation.
• Generate 1500 images with 640480 resolution

• Data filtering
• Use threshold to select the squares that have at
  least a portion of black, exclude completely
  blank squares for training

6
Experimental Results Data generation and
preprocessing (cont.)
(a)
(b)
Fig.3 (a) Original synthesized 640 480 3D image
and (b) converted 6464 2D image. Only none white
squares are used for training. Trying to restore
the 3D image to 2D image. The 2D chessboard looks
as if rotated 45?.
7
Experimental Results Train ANN
Fig.4 Training using GD with variable learning
rate
Fig.5 Training using GD with momentum
Recognition rate 72, GD with variable learning
rate 1.05 No significant recognition rate
difference between the GD, GD-VLR, GD-M GD-VLR
and GD-M are faster than GD
8
Experimental Results
The more training samples, the more similar are
the training and test sets As the number of
training samples increases, the training set
accuracy and test set accuracy merge
Fig.6 Training set accuracy and test set
accuracy. GD with variable learning rate
9
Conclusion and Future Work

• Conclusion
• Variations of Feed Forward ANNs are used to
  recognize the positions of chess pieces.

• Future work
• Adjusting ANN parameters
• Size of square divisions, effect of randomizing
  the training data, learning rate, momentum, etc.
• Use Confidence measure
• Use constraints to help recognition
• Look at successive frames to find out the piece
  moved use frames of different viewpoints for one
  chessboard state, etc.

10

• Thank you! _