Machine translation has traditionally been regarded by many as the field of study of linguists and natural language processing researchers. This paper attempted to offer an alternative, non-traditional approach to turn machine translation concepts into creation and reality. It has been recognized that machine translation of Chinese is an extremely difficult task. This paper introduced the CITAC technology for the design of intelligent full-text Chinese–English translation machines. The core of the CITAC technology is the central intelligence unit (CIU) which facilitates the design. The CIU grows with experience. Parsing of Chinese sentences is based on the trio-segmentation technique. The difficulties caused by cultural differences between Chinese and English have been overcome by the linguistic canonical form (LCF) transformation and the information pattern (IP) approach. Self-correction of grammatical errors is handled by the grammar marker pattern (GMP) approach. The innovative design concepts have been successfully developed and implemented in a commercial machine known as the CITAC translator which is a PC-based software system. For more information, please visit the CITAC home page at www.citac-MT.com

This paper presents the pattern recognition approach to machine translation of chinese to english.

Progressive image transmission (PIT) is often used to reduce the transmission time of an image telebrowsing system. A side effect of the PIT is the increase of computational complexity at the viewer's site. This effect is more serious in transform domain techniques than in other techniques. Recent attempts to reduce the side effect are futile as they create another side effect, namely, the discontinuous and unpleasant image build-up. Based on a practical assumption that image blocks to be inverse transformed are generally sparse, this paper presents a method to minimize both side effects simultaneously.

Pattern recognition studies have been traditionally concerned with concrete patterns such as two-dimensional images or graphics and three-dimensional objects. Little attention has been paid to abstract patterns such as design concepts and mathematical arguments. This paper attempts to address the abstract pattern recognition problem. The author presents an approach to automatic recognition of design concepts. Entities and relations are proposed for the description of design concepts. Computer algorithms are developed to capture the concepts in the design. The method of relaxation recognition is introduced to complete the recognition of abstract patterns.< >

This paper presents an integrated image segmentation method using edge and needle map which compensates deficiencies of using either edge-based approach or region-based approach. Segmentation of an image is the first and most difficult step toward symbolic transformation of a raw image, which is essential in image understanding. In industrial applications, the task is further complicated by the ubiquitous presence of specularity in most industrial parts. Three images taken from three different illumination directions were used to separate specular and Lambertian components in the images. Needle map is generated from Lambertian component images using photometric stereo technique. In one channel, edges are extracted and linked from the averaged Lambertian images providing one source of segmentation. The other channel, Gaussian curvature and mean curvature values are estimated at each pixel from least square local surface fit of needle map. Labeled surface type image is then generated using the signs of Gaussian and mean curvatures, where one of ten surface types is assigned to each pixel. Connected regions of identical surface type pixels provide the first level grouping, a rough initial segmentation. Edge information and initial segmentation of surface type are fed to an integration module which interprets the edges and regions in a consistent way. During interpretation regions are merged or split, edges are discarded or generated depending upon global surface fit error and consistency with neighboring regions. The output of integrated segmentation is an explicit description of surface type and contours of each region which facilitates recognition, localization and attitude determination of objects in the image.

This paper presents a new design approach to knowledge-based decision support systems for fault diagnosis and control for quality assurance and productivity improvement in automated manufacturing environments. Based on the observed manifestations, the knowledge-based diagnostic system hypothesizes a set of the most plausible disorders by mimicking the reasoning process of a human diagnostician. The data integration technique is designed to generate error-free hierarchical category files. A novel approach to diagnostic problem solving has been proposed by integrating the PADIKS (Pattern-Directed Knowledge-Based System) concept and the symbolic model of diagnostic reasoning based on the categorical causal model. The combination of symbolic causal reasoning and pattern-directed reasoning produces a highly efficient diagnostic procedure and generates a more realistic expert behavior. In addition, three distinctive constraints are designed to further reduce the computational complexity and to eliminate non-plausible hypotheses involved in the multiple disorders problem. The proposed diagnostic mechanism, which consists of three different levels of reasoning operations, significantly reduces the computational complexity in the diagnostic problem with uncertainty by systematically shrinking the hypotheses space. This approach is applied to the test and inspection data collected from a PCB manufacturing operation.© (1991) COPYRIGHT SPIE--The International Society for Optical Engineering. Downloading of the abstract is permitted for personal use only.

An intelligent robot is considered as consisting of five components: mechanisms, computer planner, computer controller, sensory systems, and knowledge-base systems. This paper discusses various aspects of robotic sensing and the need for sensor science, and introduces the design of sensory knowledge base and the knowledge-based system approach to redundant and multi sensing. The top level of the knowledge base consists of sensors, algorithms, processor, integration, and analysis. The goals for redundant and multi sensing are explained. The architecture for redundant and multi sensing system is discussed. For achieving information integration in systems with redundant sensors, we suggest Boolean fusion, probabilistic fusion, and Markov renewal analysis in addition to geometrical fusion.

The authors present a normal vector equalization method for separating probable highlights and obtaining surface orientations of 3-D specular objects. Based on the simplified Torrance-Sparrow model of specular reflection and the empirical Lambertian reflection model, this method uses a set of three monochromatic images taken from three different illumination directions, separates probable highlights from those images, and generates a set of three Lambertian (highlight-free) images and a set of three specular (highlight-only) images. Surface orientations are obtained using the Lambertian images by solving an imaging equation for a Lambertian surface. Computer simulations and experiments with real objects have been performed. The theoretical framework for measuring surface parameters which represent optical characteristics of the surface is also discussed. >

The design of a solder joint inspection system which will be used as an integral part of the PCB/AID (automated inspection and diagnosis for PC board manufacturing) system is presented. The inspection system utilizes four frames of solder joint images, extracts 15 features from the images to categorize the most important seven classes of solder joint defects, and provides two forms of inspection results according to their applications (touch-up of defects and fault diagnosis for online process control). The system has been tested for several PC boards provided by electronics firms, and experimental results have shown that it has exceptional promise.<>

This paper presents the design concept of an intelligent information system for verification, validation, association, and conversion of test data which are gathered at various workcells in CIM environment. The technician at each workcell enters test data into a central database. Experience has told us that the database may be erroneous due to wrong entry, wrong unit, wrong format, wrong code, and wrong scale. Thus the database must be verified and validated before it becomes useful. The data in the validated database are then grouped and categorized for the well-defined hierarchical structure. To facilitate the use of the database, the data will be dynamically reorganized to meet the needs of various users.

An approach to computer recognition of 3-D objects is presented. The vision system acquires the useful knowledge about the objects by taking their pictures at several stable positions and with various viewing angles. The acquired knowledge is represented by four levels of geographic codes. At each stable position, the object is characterized by a set of geographic code clusters. During the training phase, a knowledge base for geographic encoding representation is created. In the recognition phase, the vision system extracts the geographic codes of unknown objects and performs pattern cluster matching for the identification of the objects