This paper describes a method that can improve the visibility of features in shadowed regions in urban aerial images while retaining nonshadowed regions and the natural tint of shadowed regions automatically. The method first calculates the posteriori probabilities of each pixel belonging to shadowed class or nonshadowed class using color features from a low-frequency component image, and then compensates the intensity and saturation of each pixel in shadowed regions according to the posteriori probabilities. Experimental results indicate the effectiveness of proposed method.

A prototype system for drawing interpretation that automatically converts hand-sketched floor-plans into the CAD format is presented. The main problems in interpreting hand-sketched drawings are positional ambiguity and shape distortion. In the proposed method, positional ambiguity is able to be corrected in a nearest grid-location manner because floor-plan elements are normally drawn on/along grid lines that correspond to the modular proportions. Shape distortion is able to be corrected by template-matching of closed regions because architectural elements mostly consist of one or more closed regions of basic figures: triangles, squares, circles, and so on. The system was tested using 150 floor-plan drawings hand-sketched by three subjects without the use of rulers or templates. The results showed that sufficient recognition performance for practical use was able to be obtained.

This paper presents a new method for extracting the 3-D shape and texture of an object undergoing translational motion from image sequences captured through a monocular extra-wide picture viewing angle. The feature of this work is extracting this information from image sequences without requiring rigid environmental conditions. In this method, the relative positions between target and view position are estimated based on spatio-temporal image analysis, and shape is reconstructed from the multiple silhouette information. After reconstructing the 3-D shape, the voxel value of a surface point is determined by statistically analyzing those images that contain the surface point. The proposed method can extract 3-D shape and surface texture at a stroke from outdoor scenes. An experiment using real outdoor scenes confirms the effectiveness of the method.>

An effective algorithm for character recognition in scene images is studied. Scene images are segmented into regions by an image segmentation method based on adaptive thresholding. Character candidate regions are detected by observing gray-level differences between adjacent regions. To ensure extraction of multisegment characters as well as single-segment characters, character pattern candidates are obtained by associating the detected regions according to their positions and gray levels. A character recognition process selects patterns with high similarities by calculating the similarities between character pattern candidates and the standard patterns in a dictionary and then comparing the similarities to the thresholds. A relaxational approach to determine character patterns updates the similarities by evaluating the interactions between categories of patterns, and finally character patterns and their recognition results are obtained. Highly promising experimental results have been obtained using the method on 100 images involving characters of different sizes and formats under uncontrolled lighting.>

A new general purpose method for object extraction and detection, RONPaC (Robust Object Extraction (Detection) using NPaC Features) method is presented. RONPaC employs normalized principal component (NPaC) features as a measure of similarity between corresponding regions of a target image and a background image. No a priori knowledge of objects and no assumptions about the environment are required. The object extraction problem is dealt with as a discriminant problem of two classes, 'object' and 'background', in the feature space. The performance of the method is quantitatively evaluated using various real images and compared with conventional methods using two criteria, separability between two classes in feature space and ease of binarizing expressed by the maximum discriminant criterion. Experimental results confirm the extraction accuracy and applicability of the proposed method.>

An automatic and dynamic thresholding algorithm based on an illumination-independent contrast measure is proposed. In this algorithm, a local threshold is calculated for each segmented square region in an image. Since the thresholds obtained in regions that include no characters or objects would produce only noise by thresholding, they are modified by regionwise interpolation based on the contrast measure. The contrast measure is introduced by an illumination reflectance image formation model proposed by T.G. Stockham (1972) where the gray-level function is expressed as the product of an illumination component and a reflectance component. The contrast measure is expressed by the ratio of the standard derivation to the mean of the object reflectance. It is found that, if the illumination function is smooth, the ratio can be calculated from just an image function and the characteristics of the video camera without knowing about the shape of the illumination function. The characteristics are determined for two types of video cameras. The experimental results using 100 outdoor scene images for each camera show that (1) almost all the characters included in the images successfully binarized with the same parameters by this method; (2) this algorithm is independent of character or object size and its gray-level values; (3) the contrast measure is approximately independent of illumination; and (4) this method is applicable to real-world scene images.>

An effective extraction algorithm for character recognition in scene images is developed. Character candidate regions are detected by an image segmentation method based on adaptive thresholding and by evaluating gray-level difference between adjacent regions. Character pattern candidates are obtained by associating the detected regions according to their positions and gray levels and evaluating the aspect ratio of the rectangle that circumscribes the associated regions. The character recognition process selects character pattern candidates with high similarities to any categories in a dictionary (high similarity patterns). Highly promising experimental results have been obtained using the method on 100 images acquired under uncontrolled lighting. This algorithm can be applied to read characters of any size and format under various lighting conditions.>