Huibao Lin

Orthogonal Rotation-Invariant Moments for Digital Image Processing

Orthogonal rotation-invariant moments (ORIMs), such as Zernike moments, are introduced and defined on a continuous unit disk and have been proven powerful tools in optics applications. These moments have also been digitized for applications in digital image processing. Unfortunately, digitization compromises the orthogonality of the moments and, therefore, digital ORIMs are incapable of representing subtle details in images and cannot accurately reconstruct images. Typical approaches to alleviate the digitization artifact can be divided into two categories: (1) careful selection of a set of pixels as close approximation to the unit disk and using numerical integration to determine the ORIM values, and (2) representing pixels using circular shapes such that they resemble that of the unit disk and then calculating ORIMs in polar space. These improvements still fall short of preserving the orthogonality of the ORIMs. In this paper, in contrast to the previous methods, we propose a different approach of using numerical optimization techniques to improve the orthogonality. We prove that with the improved orthogonality, image reconstruction becomes more accurate. Our simulation results also show that the optimized digital ORIMs can accurately reconstruct images and can represent subtle image details.

Knowledge-based hierarchical region-of-interest detection

Detecting regions of interest (ROIs) in a complex image is a critical step in many image processing applications. In this paper, we present a new algorithm that addresses several challenges in ROI detection. The novelty of our algorithm includes: (i) every ROI contains one and only one object; (ii) the detected ROIs can have irregular shapes as opposed to the rectangular shapes that are typical of other algorithms; (iii) the algorithm is applicable to images that contain connected objects, or when the objects are broken into pieces; (iv) the algorithm is not sensitive to contrast levels in the image, and is robust to noise. These characteristics make the proposed algorithm applicable to low-resolution, real-world imagery without costly post-processing. The proposed algorithm is shown to provide outstanding performance with low-quality imagery, and is shown to be fast and robust.

Dynamic point selection in image mosaicking

Image mosaicking is a procedure of integrating information from a series of images to create a comprehensive view of the scene. It is typically carried out by selecting a subset of pixels from each of the individual images, matching these selected pixels from different images, and then mapping all the images onto a common image grid. The number of selected pixels is a critical parameter that affects both computational complexity and mosaicking accuracy. An image mosaicking algorithm is developed by using a novel dynamic point selection concept. The algorithm automatically determines the number of pixels to select according to the similarity of the images. Simulations show that the proposed algorithm generates mosaic accurately and efficiently.

Hierarchical region-of-interest detection

The objective of region-of-interest (ROI) detection is to identify portions of an image that contain targets. It is an important procedure for reducing data volume without losing important target information. The proposed algorithm aims at generating accurate representations of the ROIs such that each ROI covers one complete target. In addition, computation of the proposed algorithm is reduced by making use of information such as image resolution and approximate size of the targets. For this purpose, a multiscale, hierarchical architecture is developed. Results of the ROI detection in low-quality aerial imagery are provided to illustrate the effectiveness of the proposed algorithm.

Region-of-Interest Detection and its Application to Image Segmentation and Compression

In many intelligence, surveillance, and reconnaissance (ISR) applications, it is desirable to detect potential targets or regions of interest (ROIs) within various types of digital imagery. These ROIs can then be used to automatically cue sensors and assets, to control intelligent region-of-interest-based image compression algorithms, and to direct further analysis for target identification and recognition. In this article, we present our hierarchical region-of-interest detection (HROID) algorithm that can quickly identify and isolate critical regions within low-quality imagery

Pattern recognition and image reconstruction using improved digital Zernike moments

Zernike moments are one of the most effective orthogonal, rotation-invariant moments in continuous space. Unfortunately, the digitization process necessary for use with digital imagery results in compromised orthogonality. In this work, we introduce improved digital Zernike moments that exhibit much better orthogonality, while preserving their inherent invariance to rotation. We then propose a novel pattern recognition algorithm that is based on the improved digital Zernike moments. With the improved orthogonality, targets can be represented by fewer moments, thus minimizing computational complexity. Additionally, the rotation invariance enables our algorithm to recognize targets with arbitrary orientation. Because our algorithm eliminates the segmentation step that is typically applied in other techniques, it is better suited to low-quality imagery. Simulations on real images demonstrate these aspects of the proposed algorithm.

Fast watershed aided by target detection

The Watershed algorithm has been studied extensively, and has been applied to image segmentation due to its accuracy and robustness. However, the watershed requires a large amount of memory, and is computationally intractable for segmenting large images. In this paper, we introduce a novel hierarchical region-of-interest (ROI) detection scheme, which is used as a prelude to segmentation. With the help of the detection algorithm, watershed segmentation can be applied to the small detected regions, rather than to the entire image. Therefore, it can process large images by selectively segmenting ROIs. We focus on our new ROI detection algorithm, and how it is integrated into a system for large-image segmentation. We demonstrate the efficiency of the proposed scheme by processing a variety of images.

Fast and robust image mosaicking for monocular video

Image mosaicking is the process of mapping an image series onto a common image grid, where the resulting mosaic forms a comprehensive view of the scene. This paper presents a near-real-time, automatic image mosaicking system that is designed to operate in real-world conditions. These conditions include arbitrary camera motion, disturbances from moving objects and annotations, and luminance variations. In the proposed algorithm, matching filters are used in conjunction with automatic corner detection to find several critical points within each image, which are then used to represent the image efficiently and accurately. Numerical techniques are used to distinguish between those points belonging to the actual scene and those resulting from a disturbance, and to determine the movement of the camera. The affine model is used to describe the frame-to-frame differences that result from camera motion. A local-adaptive fine-tuning step is used to correct the approximation error due to the use of the affine model, and to compensate for any luminance variation. The mosaic is constructed progressively as new images are being added. The proposed algorithm has been extensively tested on real-world, monocular video sequences, and it is shown to be very accurate and robust.

Migrating orthogonal rotation-invariant moments from continuous to discrete space [image processing applications]

Orthogonality and rotation invariance are important feature properties in digital signal processing. Orthogonality enables a target to be represented by a compact number of features, while rotation invariance results in unique features for a target with different orientations. The orthogonal, rotation-invariant moments (ORIMs), such as Zernike, pseudo-Zernike, and orthogonal Fourier-Melling moments, are defined in continuous space. These ORIMs have been digitized and have been demonstrated effectively for some digital imagery applications. However, digitization compromises the orthogonality of the moments, and hence, reduces their precision. Therefore, digital ORIMs are incapable of representing the fine details of images. In this paper, we propose a numerical optimization technique to improve the orthogonality of the digital ORIMs. Simulation results show that our optimized digital ORIMs can be used to reproduce subtle details of images.

Automatic target detection in UAV imagery using image formation conditions

This paper is about automatic target detection (ATD) in unmanned aerial vehicle (UAV) imagery. Extracting reliable features under all conditions from a 2D projection of a target in UAV imagery is a difficult problem. However, since the target size information is usually invariant to the image formation proces, we propose an algorithm for automatically estimating the size of a 3D target by using its 2D projection. The size information in turn becomes an important feature to be used in a knowledge-driven, multi-resolution-based algorithm for automatically detecting targets in UAV imagery. Experimental results show that our proposed ATD algorithm provides outstanding detection performance, while significantly reducing the false alarm rate and the computational complexity.