Research Article
Urdu Handwritten Characters Data Visualization and Recognition Using Distributed Stochastic Neighborhood Embedding and Deep Network
Figure 7
Standard (t) -SNE based visualization results of the combined data of pixel- and structure-based information of the Urdu handwritten numerals. (a) Perplexity: 30 Iterations: 1,000. (b) Perplexity: 50 Iterations: 1,000. (c) Perplexity: 70 Iterations: 1,000. (d) Perplexity: 100 Iterations: 1,000.
(a) |
(b) |
(c) |
(d) |