Akihiro Minagawa

Local Consistency Constrained Adaptive Neighbor Embedding for Text Image Super-Resolution

This paper proposes a robust single-image super-resolution method for enlarging low quality camera captured text image. The contribution of this work is twofold. First, we point out the non-local reconstruction problem in neighbor embedding based super-resolution by statistical analysis on an empirical data set. Second, we introduce a local consistency constraint to explicitly regularize the linear reconstruction process, and adaptively generate the most possible candidates for the high-resolution image patch. For the non-consistent candidates, we rely on its adjacent overlapping patches for capability verification. Experimental results demonstrate that our solution produces visually pleasing enlargements for various text images.

A Dual Pass Video Stabilization System Using Iterative Motion Estimation and Adaptive Motion Smoothing

In this paper, we propose a novel dual pass video stabilization system using iterative motion estimation and adaptive motion smoothing. In the first pass, the transformation matrix to stabilize each frame is returned. The global motion estimation is carried out by a novel iterative method. The intentional motion is estimated using adaptive window smoothing. Before the beginning of the second pass, we obtain the optimal trim size for a specific video based on the statistics of the transformation parameters. In the second pass, the stabilized video is composed according to the optimal trim size. Experimental results show the superior performance of the proposed method in comparison to other existing methods.

A Fast Caption Detection Method for Low Quality Video Images

Captions in videos are important and accurate clues for video retrieval. In this paper, we propose a fast and robust video caption detection and localization algorithm to handle low quality video images. First, the stroke response maps from complex background are extracted by a stoke filter. Then, two localization algorithms are used to locate thin stroke and thick stroke caption regions respectively. Finally, a HOG based SVM classifier is carried out on the detected results to further remove noises. Experimental results show the superior performance of our proposed method compared with existing work in terms of accuracy and speed.

Enhancement of camera-based whiteboard images

Quality of camera-based whiteboard images is highly related to the light environment and the writing effect of the content. Specular reflection and low contrast reduce the readability of captured whiteboard images frequently. A novel method is proposed to enhance camera-based whiteboard images in this paper. The images are enhanced by removing the highlight specular reflection to improve the visibility and emphasizing the content to improve the readability of the whiteboards. The method can be practically embedded in mobile devices with image capturing cameras.

A Study on Caption Recognition for Multi-color Characters on Complex Background

We propose a caption recognition method for multicolor characters on complex background. Caption characters are used for an efficient search on a large amount of recorded TV programs. In the caption character recognition, the caption appearance section and the area is extracted, the character strokes are extracted from the area, and recognized. This paper focuses on caption character strokes extraction and recognition for multi-color characters on complex background which is a very difficult task for the conventional methods. The proposed method extracts decomposed binary images from input color caption image by color clustering. Then character candidates that are composed of combination of connect components are extracted by using recognition certainty. Finally, characters are selected by beyond-color Dynamic Programming method in which weight on recognition certainty and character alignment are used. In the character recognition evaluation of one-line multi-color character string on a complex background, a great improvement was achieved from a conventional technique that can recognize only one-color characters on complex background image.

A new pre-classification method based on associative matching method

Reducing the time complexity of character matching is critical to the development of efficient Japanese Optical Character Recognition (OCR) systems. To shorten processing time, recognition is usually split into separate preclassification and recognition stages. For high overall recognition performance, the pre-classification stage must both have very high classification accuracy and return only a small number of putative character categories for further processing. Furthermore, for any practical system, the speed of the pre-classification stage is also critical. The associative matching (AM) method has often been used for fast pre-classification, because its use of a hash table and reliance solely on logical bit operations to select categories makes it highly efficient. However, redundant certain level of redundancy exists in the hash table because it is constructed using only the minimum and maximum values of the data on each axis and therefore does not take account of the distribution of the data. We propose a modified associative matching method that satisfies the performance criteria described above but in a fraction of the time by modifying the hash table to reflect the underlying distribution of training characters. Furthermore, we show that our approach outperforms pre-classification by clustering, ANN and conventional AM in terms of classification accuracy, discriminative power and speed. Compared to conventional associative matching, the proposed approach results in a 47% reduction in total processing time across an evaluation test set comprising 116,528 Japanese character images.

Separate Chinese Character and English Character by Cascade Classifier and Feature Selection

The separation of Chinese character and English character is helpful for OCR technique. In this paper, a multi-level cascade classifier combined with feature selection is constructed to identify Chinese character and English character based on individual character. Most of samples are identified by the first node classifier, the remained low classification confidence samples are fed to the next node classifiers to get the final result. For the motivation of utilizing feature complementarity, each node classifier is trained on low classification confidence samples of its previous node classifier with independent feature selection. Furthermore, a confidence bias is utilized to improve the classifier generalization. The experiment results validate the effectiveness of this classifier.

Logical Structure Analysis for Form Images with Arbitrary Layout by Belief Propagation

A new method for analyzing the specific logical structure of forms with unknown layout is proposed. This method uses both the target form image and a generic logical structure as inputs, and models two types of relationships probabilistically: that between strings and logical components, and that between neighboring strings having different logical components. This modeling approach allows strings to be assigned to logical components softly but robustly, and allows the use of an intuitive Bayesian probability network similar to the generic logical structure. Based on this probability network model, strings corresponding to logical components can be determined by belief propagation. This method is demonstrated to be effective by conducting tests on three types of forms.

A Fast and Robust Multi-color Object Detection Method with Application to Color Chart Detection

In this paper, we focus on robust multi-color object detection with cluttered backgrounds and variable illumination for a target application to color chart detection. The task is characterized by a wide range of color variation combined with complex background. Arbitrary placement of the chart in the scene will further complicate the detection task. Conventional methods to this problem normally give an approximate bounding box of the detection result, lacking in an internal geometrical representation. Our method adopts a coarse-to-fine strategy to predict the chart location and recover its accurate topological structure, e.g. the position and boundary of each constituent color area. With this fine detection result, color deviation in the input image can be easily corrected using off-the-shelf softwares such as Photoshop. Experiential results on a public dataset demonstrate that our system can work effectively in real time and give a superior detection rate to the state-of-art. The robustness of this method to large color distortion makes it equally applicable to detection of general multi-color object such as address plate, traffic sign, and so on.

Hybrid gesture recognition system for short-range use

In recent years, various gesture recognition systems have been studied for use in television and video games[1]. In such systems, motion areas ranging from 1 to 3 meters deep have been evaluated[2]. However, with the burgeoning popularity of small mobile displays, gesture recognition systems capable of operating at much shorter ranges have become necessary. The problems related to such systems are exacerbated by the fact that the cameras field of view is unknown to the user during operation, which imposes several restrictions on his/her actions. To overcome the restrictions generated from such mobile camera devices, and to create a more flexible gesture recognition interface, we propose a hybrid hand gesture system, in which two types of gesture recognition modules are prepared and with which the most appropriate recognition module is selected by a dedicated switching module. The two recognition modules of this system are shape analysis using a boosting approach (detection-based approach)[3] and motion analysis using image frame differences (motion-based approach)(for example, see[4]). We evaluated this system using sample users and classified the resulting errors into three categories: errors that depend on the recognition module, errors caused by incorrect module identification, and errors resulting from user actions. In this paper, we show the results of our investigations and explain the problems related to short-range gesture recognition systems.