Description: We present a video caption detection and recognition
system based on a fuzzy-clustering neural network (FCNN) classifier.
Using a novel caption-transition detection scheme we locate
both spatial and temporal positions of video captions with high precision
and efficiency. Then employing several new character segmentation
and binarization techniques, we improve the Chinese
video-caption recognition accuracy from 13 to 86 on a set of
news video captions. As the first attempt on Chinese video-caption
recognition, our experiment results are very encouraging.-A spatial-temporal approach for video caption date Platform: |
Size: 466944 |
Author:段军伟 |
Hits:
Description: Research on Video Caption Detection and Extraction. The automatic video annotation is a key module of the video indexing and retrieval system,and the extraction and location of captions are important steps for video annotation.An automatic extraction algorithm for video caption by combining the wavelet transform and the fuzzy clustering neural network in presented.This algorithm is especially applicable to extracting and locating of Chinese captions.The theoretical reachs show that the proposed algorithm can achieve an accuracy in more than 90 in caption detection and location.After the selected characters in the region to identify with OCR technology,OCR recognition accuracy can reach 99 .-Research on Video Caption Detection and Extraction. The automatic video annotation is a key module of the video indexing and retrieval system,and the extraction and location of captions are important steps for video annotation.An automatic extraction algorithm for video caption by combining the wavelet transform and the fuzzy clustering neural network in presented.This algorithm is especially applicable to extracting and locating of Chinese captions.The theoretical reachs show that the proposed algorithm can achieve an accuracy in more than 90 in caption detection and location.After the selected characters in the region to identify with OCR technology,OCR recognition accuracy can reach 99 . Platform: |
Size: 13312 |
Author:段军伟 |
Hits:
Description: RKButton是什么？简单来说就是具有vista按钮一切效果的按钮控件
抽空更新了一下，抛弃了label，改用drawtext画caption，于是解决了闪烁问题。也很完美的搞定了drawtext自动换行以及caption的垂直居中定位问题（算法可能还不够快，但好用就行了。。仅在caption change和resize的时候重新计算一下），加入了空格键的检测，现在这个按钮的行为模式和标准commandbutton应该都一样了。
另外优化了几个地方，具体就不写了-RKButton What is this? Is simply all the results with the vista button, the button controls
Taking the time to update a moment abandoned the label, use drawtext picture caption, then solve the flicker problem. Is also the perfect Gaoding the drawtext wrap, as well as the vertical center positioning of caption (algorithm might not be fast enough, but it handy on the list.. Only in the caption change and resize the time to re-calculate), joined the space bar detection Now the button patterns of behavior and standards of commandbutton should all the same.
Also optimized in several places, specifically not to write a Platform: |
Size: 8192 |
Author:黄雯琼 |
Hits:
Search - caption detection