ANGY: A Rule-Based Expert System for Automatic Segmentation of Coronary Vessels From Digital Subtracted Angiograms

This paper details the design and implementation of ANGY, a rule-based expert system in the domain of medical image processing. Given a subtracted digital angiogram of the chest, ANGY identifies and isolates the coronary vessels, while ignoring any nonvessel structures which may have arisen from noise, variations in background contrast, imperfect subtraction, and irrelevent anatomical detail. The overall system is modularized into three stages: the preprocessing stage and the two stages embodied in the expert itself. In the preprocessing stage, low-level image processing routines written in C are used to create a segmented representation of the input image. These routines are applied sequentially. The expert system is rule-based and is written in OPS5 and LISP. It is separated into two stages: The low-level image processing stage embodies a domain-independent knowledge of segmentation, grouping, and shape analysis. Working with both edges and regions, it determines such relations as parallel and adjacent and attempts to refine the segmentation begun by the preprocessing. The high-level medical stage embodies a domain-dependent knowledge of cardiac anatomy and physiology. Applying this knowledge to the objects and relations determined in the preceding two stages, it identifies those objects which are vessels and eliminates all others.     

The interpretation of laser radar images by a knowledge-based system

This paper presents a knowledge-based system to interpret laser radar (ladar) images. The objective of this research is to detect and recognize man-made objects in outdoor scenes. Our system applies themultisensor fusion approach to multiple ladar modalities to improve both segmentation and interpretation. The segmentation modules are written in C. The knowledge-based interpretation system is constructed usingKEE and Lisp. Low-level attributes of image segments (regions) are computed by the segmentation modules and then converted to theKEE format. The interpretation system applies forward chaining in a bottom-up fashion to derive object-level interpretation from input generated by low-level processing and segmentation modules. The interpretation modules detect man-made objects from the background using low-level attributes. Segments are grouped into objects, which are then classified into predefined categories (vehicles, ground, etc.). The efficiency of the interpretation system is enhanced by transferring nonsymbolic processing tasks to a concurrent service manager (program). Experimental results using ladar data are presented.     

Unsupervised texture segmentation using Gabor filters

This paper presents a texture segmentation algorithm inspired by the multi-channel filtering theory for visual information processing in the early stages of human visual system. The channels are characterized by a bank of Gabor filters that nearly uniformly covers the spatial-frequency domain, and a systematic filter selection scheme is proposed, which is based on reconstruction of the input image from the filtered images. Texture features are obtained by subjecting each (selected) filtered image to a nonlinear transformation and computing a measure of “energy” in a window around each pixel. A square-error clustering algorithm is then used to integrate the feature images and produce a segmentation. A simple procedure to incorporate spatial information in the clustering process is proposed. A relative index is used to estimate the “true” number of texture categories.     

A framework for heading-guided recognition of human activity

A combined 2D, 3D approach is presented that allows for robust tracking of moving people and recognition of actions. It is assumed that the system observes multiple moving objects via a single, uncalibrated video camera. Low-level features are often insufficient for detection, segmentation, and tracking of non-rigid moving objects. Therefore, an improved mechanism is proposed that integrates low-level (image processing), mid-level (recursive 3D trajectory estimation), and high-level (action recognition) processes. A novel extended Kalman filter formulation is used in estimating the relative 3D motion trajectories up to a scale factor. The recursive estimation process provides a prediction and error measure that is exploited in higher-level stages of action recognition. Conversely, higher-level mechanisms provide feedback that allows the system to reliably segment and maintain the tracking of moving objects before, during, and after occlusion. Heading-guided recognition (HGR) is proposed as an efficient method for adaptive classification of activity. The HGR approach is demonstrated using “motion history images” that are then recognized via a mixture-of-Gaussians classifier. The system is tested in recognizing various dynamic human outdoor activities: running, walking, roller blading, and cycling. In addition, experiments with real and synthetic data sets are used to evaluate stability of the trajectory estimator with respect to noise.     

3-D shape recovery using distributed aspect matching

An approach to the recovery of 3-D volumetric primitives from a single 2-D image is presented. The approach first takes a set of 3-D volumetric modeling primitives and generates a hierarchical aspect representation based on the projected surfaces of the primitives; conditional probabilities capture the ambiguity of mappings between levels of the hierarchy. From a region segmentation of the input image, the authors present a formulation of the recovery problem based on the grouping of the regions into aspects. No domain-independent heuristics are used; only the probabilities inherent in the aspect hierarchy are exploited. Once the aspects are recovered, the aspect hierarchy is used to infer a set of volumetric primitives and their connectivity. As a front end to an object recognition system, the approach provides the indexing power of complex 3-D object-centered primitives while exploiting the convenience of 2-D viewer-centered aspect matching; aspects are used to represent a finite vocabulary of 3-D parts from which objects can be constructed     

Color-Based Image Salient Region Segmentation Using Novel Region Merging Strategy

In this paper, we propose a novel unsupervised algorithm for the segmentation of salient regions in color images. There are three phases in this algorithm. In the first phase, we use nonparametric density estimation to extract candidates of dominant colors in an image, which are then used for the quantization of the image. The label map of the quantized image forms initial regions of segmentation. In the second phase, we define salient region with two properties; i.e., conspicuous; compact and complete. According to the definition, two new parameters are proposed. One is called “Importance index”, which is used to measure the importance of a region, and the other is called “Merging likelihood”, which is utilized to measure the suitability of region merging. Initial regions are merged based on the two new parameters. In the third phase, a similarity check is performed to further merge the surviving regions. Experimental results show that the proposed method achieves excellent segmentation performance for most of our test images. In addition, the computation is very efficient.      

FILIP: A fuzzy intelligent information system with learning capabilities

FILIP (fuzzy intelligent learning information processing) system is designed with the goal to model human information processing. The issues addressed are uncertain knowledge representation and approximate reasoning based on fuzzy set theory, and knowledge acquisition by “being told” or by “learning from examples”. Concepts that can be “learned” by the system can be imprecise (fuzzy), or the knowledge can be incomplete. In the latter case, FILIP uses the concept of similarity to extrapolate the knowledge to cases that were not covered by examples provided by the user. Concepts are stored in the Knowledge Base and employed in intelligent query processing, based on flexible inference that supports approximate matches between the data in the database and the query.

The architecture of FILIP is discussed, the learning algorithm is described, and examples of the system's performance in the knowledge acquisition and querying modes, together with its explanatory capabilities are shown.     

改进Mask R-CNN网络在医学图像识别与分割中的应用

针对现有医学图像处理方法在人体复杂结构组织器官分割中的不足,提出复用低层特征信息的Mask R-CNN网络。该网络可对特定组织器官识别时同时进行分割,为了提高包含较多细节信息的低层特征层的利用率,将低层的特征信息添加到高层的特征中,使低层与高层特性优劣互补,将原始图像首次长宽压缩两次后的特征层定义为C1层,而后分别通过复用C1层和复用依次卷积的C1层这两种方法实现。并将主干网络进行了精简,以加快网络的训练速度,降低识别和分割的时间。以下颌骨作为应用对象,自建包含1?064张下颌骨CT图片的数据集,按9∶1的比例划分为训练集和验证集进行训练,使得复用依次卷积C1层的Mask R-CNN网络的训练损失降至2.8%,验证损失降至6.6%,表明该网络在下颌骨的识别和分割上具有很高的准确率。     

A generic knowledge-guided image segmentation and labeling system using fuzzy clustering algorithms

Segmentation of an image into regions and the labeling of the regions is a challenging problem. In this paper, an approach that is applicable to any set of multifeature images of the same location is derived. Our approach applies to, for example, medical images of a region of the body; repeated camera images of the same area; and satellite images of a region. The segmentation and labeling approach described here uses a set of training images and domain knowledge to produce an image segmentation system that can be used without change on images of the same region collected over time. How to obtain training images, integrate domain knowledge, and utilize learning to segment and label images of the same region taken under any condition for which a training image exists is detailed. It is shown that clustering in conjunction with image processing techniques utilizing an iterative approach can effectively identify objects of interest in images. The segmentation and labeling approach described here is applied to color camera images and two other image domains are used to illustrate the applicability of the approach.     

A Dual Source,Parallel Architecture for Computer Vision

We present a parallel architecture for object recognition and location based on concurrent processing of depth and intensity image data. Parallel algorithms for curvature computation and segmentation of depth data into planar or curved surface patches, and edge detection and segmentation of intensity data into extended linear features, are described. Using this feature data in comparison with a CAD model, objects can be located in either depth or intensity images by a parallel pose clustering algorithm.The architecture is based on cooperating stages for low/intermediate level processing and for high level matching. Here, we discuss the use of individual components for depth and intensity data, and their realisation and integration within each parallel stage. We then present an analysis of the performance of each component, and of the system as a whole, demonstrating good parallel execution from raw image data to final pose.     

New methods for matching 3-d objects with single perspective views

In this paper we analyze the ability of a computer vision system to derive properties of the three-dimensional (3-D) physical world from viewing two-dimensional (2-D) images. We present a new approach which consists of a model-based interpretation of a single perspective image. Image linear features and linear feature sets are backprojected onto the 3-D space and geometric models are then used for selecting possible solutions. The paper treats two situations: 1) interpretation of scenes resulting from a simple geometric structure (orthogonality) in which case we seek to determine the orientation of this structure relatively to the viewer (three rotations) and 2) recognition of moderately complex objects whose shapes (geometrical and topological properties) are provided in advance. The recognition technique is limited to objects containing, among others, straight edges and planar faces. In the first case the computation can be carried out by a parallel algorithm which selects the solution that has received the largest number of votes (accumulation space). In the second case an object is uniquely assigned to a set of image features through a search strategy. As a by-product, the spatial position and orientation (six degrees of freedom) of each recognized object is determined as well. The method is valid over a wide range of perspective images and it does not require perfect low-level image segmentation. It has been successfully implemented for recognizing a class of industrial parts.     

Dynamic learning from multiple examples for semantic object segmentation and search

We present a novel “dynamic learning” approach for an intelligent image database system to automatically improve object segmentation and labeling without user intervention, as new examples become available, for object-based indexing. The proposed approach is an extension of our earlier work on “learning by example,” which addressed labeling of similar objects in a set of database images based on a single example. The proposed dynamic learning procedure utilizes multiple example object templates to improve the accuracy of existing object segmentations and labels. Multiple example templates may be images of the same object from different viewing angles, or images of related objects. This paper also introduces a new shape similarity metric called normalized area of symmetric differences (NASD), which has desired properties for use in the proposed “dynamic learning” scheme, and is more robust against boundary noise that results from automatic image segmentation. Performance of the dynamic learning procedures has been demonstrated by experimental results.     

Learning Task-Specific Object Recognition and Scene Understanding

In this paper, we present an approach to object recognition and scene understanding which integrates low-level image processing and high-level knowledge-based components. A novel machine learning system is presented which is used to acquire knowledge relating to a specific task. Learned feedback from high-level to low-level processes is introduced as a means of achieving robust task-specific segmentation. The system has been implemented and trained on a number of scenarios with differing tasks from which results are presented and discussed.     

基于自组织动态神经网络的图像分割

图像分割是图像处理和模式识别的重要课题,而图像特征空间聚类是图像分割的一种重要方法,认为图像的特征是图像中待分割物体表面所特有而且恒定的特征,并将图像的特征映射到某种几何空间,称为特征空间,并且假定图像中不同的待分割物体在该特征空间中呈现为不同的聚集,提出了自组织动态网络(SODNN)聚类算法,并且利用该算法对图像特征空间聚类.该算法实现了神经网络结构的快速生长和动态调节,具有自动适应数据内在分布特征和聚类结果更为准确稳定的特点.利用SODNN算法对图像颜色空间进行聚类的同时综合了图像的位置信息来实现图像分割.实验表明分割结果与人工分割结果具有较好的一致性.     

Toward an expert system for chromosome analysis

Automation of chromosome analysis has long been considered as a very difficult task. Efforts to computerize some or all of the procedures using various conventional pattern recognition techniques have had only limited success. In this paper the previous work in this domain is briefly reviewed, with a discussion of the limitations of the existing approaches. An attempt to apply AI techniques is introduced, and the organization of an expert image analysis system for chromosome classification is described, mainly at a conceptual level. Based on the proposed architecture, the low-level processes (segmentation, feature extraction) and the high-level processes (classifications or interpretations) can be carried out in a knowledge-guided fashion with a combinational use of image processing and pattern recognition knowledge, as well as expert chromosome classification knowledge embedded in a rule-based structure. A knowledge-based chromosome image analysis scheme is presented which adopts a hierarchical hypothesize-and-verify paradigm. Example rules are given to illustrate how they can be used in this scheme.     

受限领域中文文本内容主题概念识别研究

该文尝试利用领域知识库,实现受限领域中文文本内容主题概念识别,并尝试利用领域概念建立特征向量,实现受限领域文本与主题概念的相关度计算。其中,利用领域知识库进行主题特征识别,利用知识库中概念间的关系进行特征项权重的计算,并利用knn算法进行文本与主题概念的相关度计算。实现了基于词汇分析技术和基于领域知识计算相结合的方式下,受限领域中文文本内容主题概念的识别和文本与主题概念的相关度计算。实验结果显示,主题识别平均正确率为79%文本与主题概念的相关度计算的正确率达到62%。     

Segmentation and Interpretation of Multicolored Objects with Highlights

This paper presents a segmentation system, based on a general framework for segmentation, that returns not only regions that correspond to coherent surfaces in an image, but also low-level interpretations of those regions' physical characteristics. This system is valid for images of piecewise uniform dielectric objects with highlights, moving it beyond the capabilities of previous physics-based segmentation algorithms which assume uniformly colored objects. This paper presents a summary of the complete system and focuses on two extensions of it that demonstrate its interpretive capacity and applicability to more complex scenes. The first extension provides interpretations of a scene by reasoning about the likelihood of different physical characteristics of simple image regions. The second extension allows the system to handle highlights within the general framework for segmentation. The resulting segmentations and interpretations more closely match our perceptions of objects since the resulting regions correspond to coherent surfaces, even when those surfaces have multiple colors and highlights.     

Retrieval of movie scenes by semantic matrix and automatic feature weight update

A new semantic-based video scene retrieval method is proposed in this paper. Twelve low-level features extracted from a video clip are represented in a genetic chromosome and target videos that user has in mind are retrieved by the interactive genetic algorithm through the feedback iteration. In this procedure, high-level semantic relevance between retrieved videos is accumulated with so-called semantic relevance matrix and semantic frequency matrix for each iteration, and they are combined with an automatic feature weight update scheme to retrieve more target videos at the next iteration. Experiments over 300 movie scene clips extracted from latest well-known movies, showed an user satisfaction of 0.71 at the fourth iteration for eight queries such as “gloominess”, “happiness”, “quietness”, “action”, “conversation”, “explosion”, “war”, and “car chase”.     

Fuzzy segmentation for object‐based image classification

This Letter proposes an object‐based image classification procedure which is based on fuzzy image‐regions instead of crisp image‐objects. The approach has three stages: (a) fuzzification in which fuzzy image‐regions are developed, resulting in a set of images whose digital values express the degree of membership of each pixel to target land‐cover classes; (b) feature analysis in which contextual properties of fuzzy image‐regions are quantified; and (c) defuzzification in which fuzzy image‐regions are allocated to target land‐cover classes. The proposed procedure is implemented using automated statistical techniques that require very little user interaction. The results indicate that fuzzy segmentation‐based methods produce acceptable thematic accuracy and could represent a viable alternative to current crisp image segmentation approaches.