Description: 为了提取区域边界，可以对图像直接运用一阶微商算子或二阶微商算子，然后根据各像点处的微商幅值或其他附加条件判定其是否为边界点。如果图像中含有较强噪声，直接进行微商运算将会出现许多虚假边界点。-.Using the feature of distinct edge contour existing between the text and the background regions in color images, a novel text extraction algorithm—CEMA(Color-edge detection, Morphology, logic operator“AND”) is proposed in this thesis for video images. Platform: |
Size: 6144 |
Author:刘卅 |
Hits:
Description: 文本挖掘的背景
文本挖掘的过程
特征抽取
特征选择
文本分类
文本聚类
模型评价
-The background of text mining text mining process of feature extraction feature selection text classification text clustering model to evaluate the Platform: |
Size: 226304 |
Author:linda |
Hits:
Description: 一篇关于文字识别中文本抽取的英文论文，适用于复杂背景的图像。-Article on the Chinese character recognition in English of the selected papers for complex background image. Platform: |
Size: 989184 |
Author:Ivan.Ru |
Hits:
Description: The basic steps of the connected-component text extraction algorithm are given below,
and diagrammed in Figure 10. The details are discussed in the following sections.
1. Convert the input image to YUV color space. The luminance(Y) value is used for
further processing. The output is a gray image.
2. Convert the gray image to an edge image.
3. Compute the horizontal and vertical projection profiles of candidate text regions
using a histogram with an appropriate threshold value.
4. Use geometric properties of text such as width to height ratio of characters to
eliminate possible non-text regions.
5. Binarize the edge image enhancing only the text regions against a plain black
background.
6. Create the Gap Image (as explained in the next section) using the gap-filling
process and use this as a reference to further eliminate non-text regions the
output.
-The basic steps of the connected-component text extraction algorithm are given below,
and diagrammed in Figure 10. The details are discussed in the following sections.
1. Convert the input image to YUV color space. The luminance(Y) value is used for
further processing. The output is a gray image.
2. Convert the gray image to an edge image.
3. Compute the horizontal and vertical projection profiles of candidate text regions
using a histogram with an appropriate threshold value.
4. Use geometric properties of text such as width to height ratio of characters to
eliminate possible non-text regions.
5. Binarize the edge image enhancing only the text regions against a plain black
background.
6. Create the Gap Image (as explained in the next section) using the gap-filling
process and use this as a reference to further eliminate non-text regions the
output.
Platform: |
Size: 41984 |
Author:Lee Kurian |
Hits:
Description: The basic steps of the edge-based text extraction algorithm are given below
1. Create a Gaussian pyramid by convolving the input image with a Gaussian kernel
and successively down-sample each direction by half. (Levels: 4)
2. Create directional kernels to detect edges at 0, 45, 90 and 135 orientations.
3. Convolve each image in the Gaussian pyramid with each orientation filter.
4. Combine the results of step 3 to create the Feature Map.
5. Dilate the resultant image using a sufficiently large structuring element (7x7 [1])
to cluster candidate text regions together.
6. Create final output image with text in white pixels against a plain black
background. Platform: |
Size: 2048 |
Author:Lee Kurian |
Hits:
Description: object classification using recognized scene text in natural images. While the state-of-the-art relies on visual cues only, this paper is the first work which proposes to combine textual and visual cues. Another novelty is the textual cue extraction. Unlike the state-of-the-art text detection methods, we focus more on the background instead of text regions. Once text regions are detected, they are further processed by two methods to perform text recognition Platform: |
Size: 8192 |
Author:MANIKANDAN |
Hits:
Search - background text extraction