Search

There is a wide range of segmentation methods for bone structures in CT images. Many of these methods are declared as automatic, but it is not guaranteed, that the resulting segmentation labels the volume of interest correctly in any case. This work presents a technique, which assists the user with the necessary corrections of the segmentation errors. The procedure must be started manually, but the following steps are fully automatic. First, a similar, correct segmentation is selected from a database, which is used to mask the defects. Then the selected segmentation is registered onto the defect one using the diffeomorphic demons algorithm. Thereby, the region inside the mask is excluded from registration but the displacement field is interpolated. The method has been implemented and tested for segmentations of the proximal femur head, but can easily be transferred to segmentations of other bone regions.

In 3DTV research, virtual view synthesis is a key component to the technology. Depth-image-based-rendering (DIBR) is an important method to realize virtual view synthesis. However, DIBR always results in hole problems where the depth and colour values are not known. Hole-filling methods often cause other problems, such as edge-ghosting and cracks. This paper proposes an algorithm that uses the depth and colour images to address the holes. It exploits the assumption of a virtual view between two laterally aligned reference cameras. The hole-filling method is performed on the blended depth image by morphological operations, and inpainting of the holes is obtained with the position information provided by the filtered depth maps. A new interpolation method to eliminate edge-ghosting is also presented, which additionally uses a post-processing technique to improve image quality. The main novelty of this paper is the unique image blending, which is more efficient than pre-processing depth maps. It is also the first method that is using morphological closing in the depth map de-noising process. The method proposed in this paper can effectively remove holes and edge-ghosting. Experimental quantitative and qualitative results show the proposed algorithm improves quality dramatically on traditional methods.

In this paper, a semi-fragile watermarking with interpolation method is proposed to improve recovery performance with less recovery watermark capacity. The original image is 2-fold down sampled to reduce the watermark payload and DCT (Discrete Cosine Transformation) is calculated on each 4 × 4 block in the down-sampled image. The DC coefficient and first two AC coefficients in each 4 × 4 block are quantized by selected quantized steps respectively and encoded with 11 bits to generate the recovery watermark corresponding to 8 × 8 block in the original image. The recovery watermark of each 8 × 8 block is embedded in the quantized DCT coefficients of other blocks. At the recovery side, the low resolution image is firstly reconstructed by the extracted valid recovery watermark and the high resolution one is reconstructed by the interpolation method based on the low resolution one. The tampered blocks are recovered by the corresponding blocks in the high resolution image. The image inpainting method is also used to recover the tampering coincident blocks. Experimental results show that the proposed restorable semi-fragile watermarking method can achieve better recovery performance under JPEG (Joint Photographic Experts Group) compression with superior invisibility.

Panoramic image-based representation of real world environments has received much attention in virtual/augmented reality applications due to advantages such as shorter generation times, faster rendering speeds, higher photorealism, and less storage needed when compared to the 3D modeling approaches. In this paper, a novel approach is proposed for generating animated stereo panoramic images based on a single video sequence captured by a panning camera. The techniques involved include generating seamless stereo video textures, inpainting the unsatisfactory regions, and embedding video textures into stereo panoramic images. A player is also developed to provide user stereo visualization and real-time navigation of the image-based virtual environment that are featured with seamlessly-looping animated scenes. The quality of the generated animated stereo panoramic image is satisfactory for virtual reality applications and the computation time is acceptable for practical use.

Dermoscopy image analysis technology is discussed based on Chinese in this chapter. It includes four aspects: preprocessing, segmentation, feature extraction and classification. Firstly, in preprocessing stage, hair is extracted out according to the elongate of connected region, and then removed from the image by using PDE-based image inpainting technology. Secondly, a novel dermoscopy image segmentation algorithm is introduced using self-generating neural network (SGNN) combined with genetic algorithm (GA). And in the feature description stage, the features including color, texture, shape and border are extracted for the lesion object. Lastly, the model of combined neural network classifier is employed to classify the lesion object successfully with a sensitivity and specificity of 93.3 and 96.7 % respectively. Based on the image analysis method disscussed in this chapter, an automatic analysis system of dermoscopy images of Chinese is successfully developed and has been applied for the clinical diagnosis of skin tumors.

In this paper, the problem of automatic people removal from digital photographs is addressed. Removing unintended people from a scene can be very useful to focus further steps of image analysis only on the object of interest, A supervised segmentation algorithm is presented and tested in several scenarios.

The efficient compression of depth maps is becoming more and more important. We present a novel codec specifically suited for this task. In the encoding step we segment the image and extract between-pixel contours. Subsequently we optimise the grey values at carefully selected mask points, including both hexagonal grid locations as well as freely chosen points. We use a chain code to store the contours. For the decoding we apply a segment-based homogeneous diffusion inpainting. The segmentation allows parallel processing of the individual segments. Experiments show that our compression algorithm outperforms comparable methods such as JPEG or JPEG2000, while being competitive with HEVC (High Efficiency Video Coding).

In this paper we present a fast video inpainting technique to infer the unknown information in the target region by maximizing box-based self-similarity and coherence measure. The video inpainting is already proposed in the literature and some of them are able to produce good quality results. However, the bottleneck of those algorithms is they are painfully slow. Here we fill the texture in the target region that preserves the smooth motion of the object without inclusion of any artifacts in reasonable amount of time. Our experiments show that the proposed method is quite efficient to synthesize unknown information in a video and comparable to the existing state-of-the-art methods. Moreover, proposed method is based on box filling and optimization is done on multiple scale using EM algorithm, and is computationally faster than the existing ones.

Historical monuments are considered as one of the key aspects for modern communities. Unfortunately, due to a variety of factors the monuments get damaged. One may think of digitally undoing the damage to the monuments by inpainting, a process to fill-in missing regions in an image. A majority of inpainting techniques reported in the literature require manual selection of the regions to be inpainted. In this paper, we propose a novel method that automates the process of identifying the damage to visually dominant regions viz. eyes, nose and lips in face image of statues, for the purpose of inpainting. First, a bilateral symmetry based method is used to identify the eyes, nose and lips. Textons features are then extracted from each of these regions in a multi-resolution framework to characterize both the regular and irregular textures. These textons are matched with those extracted from a training set of true vandalized and non-vandalized regions, in order to classify the region under consideration. If the region is found to be vandalized, the best matching non-vandalized region from the training set is used to inpaint the identified region using the Poisson image editing method. Experiments conducted on face images of statues downloaded from the Internet, give promising results.

Recently, as 3D display devices become popular, automatic editing methods for stereoscopic image become important. In this paper, we propose an inpainting method for a stereoscopic image considering the texture similarity. By extending the conventional image inpainting method for an image considering the texture similarity to for a stereoscopic image by considering the consistency of two images, natural results are obtained by our method. As in the conventional methods for a stereoscopic image inpainting, first the depth maps are recovered and then the color images are inpainted based on the depth maps in our method. Consistency of the depth maps is guaranteed in our method which are important for the inpainting of color image. For color image inpainting, similar textures are searched considering not only the depth layer but also the depth values of respective pixels. Experimental results demonstrate the efficiency of our method.

The Microsoft Kinect depth sensor provides depth maps of indoor environments with unknown depth gaps because of infrared (IR) reflectance properties of the objects inside the scene and occlusion. In this paper, we propose a grid-based hierarchical learning algorithm for predicting the depth values of the gaps inside the Kinect's depth map. The scene is divided into hierarchical grids and the depth of each grid is modeled using supervised learning. The learned models can be directly applied to upcoming frames without repeating the learning procedure and occluded regions in the background can be recovered. The proposed algorithm outperforms an inpainting-based approach quantitatively and successfully recovers the objects that are occluded at the background.

In this paper we present a novel view synthesis method for mobile platform. A disparity-based view interpolation is proposed to synthesize the virtual view from a left and right image pair. This makes to full use of the available information of both images to give accurate interpolation results and greatly decreases the pixel number in the disoccusion region, thus reducing the errors introduced by the patch-based image inpainting. Two boundary refining schemes considering gradient and color coherence and applying directional filters on boundaries are proposed to improve the synthesized results. Experimental results show that the proposed method is effective and suitable for mobile environment.

We are often required to retouch images in order to improve their visual appearance, by removing the visual discontinuities like breaks and damaged regions. Such retouching of images may be achieved by inpainting. Current techniques for image inpainting require the user to manually select the target regions to be inpainted. Very few techniques for automatically detecting the target regions for inpainting are reported in the literature, which are suitable to detect an actual damage or alteration to the given photograph. In this paper, we propose a Singular Value Decomposition (SVD) based novel technique for automatic detection of the damaged regions in the photographed object / scene, for the purpose of digitally restoring them to their entirety using inpainting. Results on an exhaustive set of images suggest that the mask generated using the proposed technique can be suitably used for inpainting purpose to digitally restore the given images.

Double-sided DIBR can generate a perfect virtual view image. However, most of the data are the same in the two candidates. It is a waste cost. This paper proposed a simpler and more effective method to reduce the cost called forward and reverse wrapping DIBR (FR-DIBR). The FR-DIBR utilizes a main color image with associated depth map and an assistant color image. First, it wraps the main reference image to the virtual plane, and then reversely wraps the region of the disocclusion in the virtual view to the assistant reference image to get the lost information of the holes. The efficiency is improved by only wrapping one image and the disocclusion region. A depth map inpainting method called BBWF is also proposed. The experimental results show that the FR-DIBR can reduce the computation cost by 33 % and generate a clearer virtual image compared with the double-sided DIBR.