This paper proposes a novel method for predicting the Remaining Useful Life (RUL) of turbine engines by integrating Convolutional Neural Networks (CNN) and Bidirectional Long Short-Term Memory (BiLSTM) frameworks with Chaotic Mapping (CM) and Fractional-Order (FO) filters (Called CBCF). The method follows a sequence of CM, FO filter, CNN, and BiLSTM. CM is used for preprocessing to extract meaningful features from engine data, while the FO filter enhances image details produced by CM, contributing to more accurate RUL predictions. The enhanced images are then used as inputs to the CNN-BiLSTM model. CNN is employed to automatically extract and learn hierarchical features from the engine data through convolutional layers, capturing spatial patterns and local dependencies for more effective feature representation. BiLSTM is then used to process the time-sequenced data, integrating forward and backward network information to capture both past and future dependencies, thereby improving prediction accuracy. The C-MAPSS dataset, which is a benchmark dataset in this field, is utilized to evaluate the feasibility and effectiveness of the proposed method. The CMAPSS dataset provides comprehensive data on engine degradation and has been widely used for validating RUL prediction models. Experimental results show that the integration of CM and FO filters with the CNN-BiLSTM model enhances performance, significantly reducing the Root Mean Square Error (RMSE) and demonstrating the potential of this approach for accurate RUL predictions.

In this paper, a deep learning (DL)-based detector, namely SSOR-Net, is proposed for massive MIMO systems. SSOR-Net enhances the conventional Symmetric Successive Over-Relaxation (SSOR) method while maintaining low complexity. These vectors are optimized through network training, using DL techniques to maximize the detector's effectiveness. Extensive simulations and complexity analysis reveal that SSOR-Net achieves performance comparable to the optimal OAMP-Net while offering significantly lower complexity. Furthermore, SSOR-Net outperforms MMSE detection, traditional SSOR methods, and other DL-based schemes, particularly in scenarios where the number of transmit antennas approaches that of the receive antennas.

In this paper we present a novel unsupervised feature representation by extracting salient symmetries in RGB-D images using the proposed moment-based symmetric patch detector. A fast indexing structure is also derived to group local symmetric patches into semantically meaningful symmetric parts. Given an RGB-D image, the hash-based symmetric patch indexing speeds up the searches of symmetric patch pairs, which are further grouped into symmetric parts with nearly linear time complexity. In the context of symmetry matching and scene classification, the second part of this work presents a symmetry-based scene modeling, aiming at computing a robust part-based feature set for each image category. To verify the effectiveness of the symmetry detector, based on the pre-learned part-based scene model, a part-based voting scheme is constructed to annotate the scene type of the input RGB-D image. Experimental results show that the proposed approach outperforms the compared methods in terms of detection and recognition accuracy using publicly available datasets.

In this paper, we present an indexing structure of data-driven cuboid patterns to speed up the process of liver detection and segmentation from computed tomography (CT) scans using the cube-based generalized Hough transform (CGHT). Most existing approaches to automatic liver segmentation from CT scans use a statistical shape model (SSM) integrated with a searching algorithm to recover the deformation. However, establishing the correspondences among landmark points of training shapes for the construction of the average shape of SSM remains as a challenge due to the high variation of liver shapes in CT scans. The proposed method is a fully automatic segmentation method that combines four steps. Firstly, a test CT volume is partitioned into multiple non-overlapped sub-volumes with each of them consisting of variable numbers of consecutive slices. Secondly, we locate the cube-based liver shapes in all sub-volumes via Hough voting and dynamic programming. Thirdly, we construct the basic 3D model through a level-set framework for liver shape segmentation. Finally, to introduce neighbors statistical analysis into the above model, we deform the 3D liver shape to overcome disturbances caused by noise and inhomogeneity. The MICCAI 2007 liver segmentation challenge datasets are used to verify the effectiveness of the proposed method. Experimental results demonstrate the good performance of the proposed method in terms of segmentation accuracy and execution speed.

This paper proposes a novel object classification method which uses the concept of probabilistic latent semantic analysis (pLSA) to overcome the problem of sparse representation in data classification. Sparse representation is widely used and quite successful in many vision-based applications. However, it needs to calculate the sparse reconstruction cost (SRC) of each sample to find the best candidate. Because an optimization process is involved, it is very inefficient. In addition, it uses only the residual and does not consider the arrangement (or distribution) of combination coefficients of visual codes in classification. Thus, it often fails to classify categories if they are similar. In this paper, the pLSA concept is first introduced into the sparse representation to build a new classifier without using the SRC measure. The weakness of the pLSA scheme is the use of EM algorithm for updating the posteriori probability of latent class. Because it is very time-consuming, a novel weighting voting strategy is introduced to improve the pLSA scheme for recognizing objects in real time. The advantages of this classifier are: the accuracy is much higher than the SRC scheme and the efficiency is real-time in data classification. Two applications are demonstrated in this paper to prove the superiority of the new classifier, i.e., vehicle make and model recognition, and action analysis.

SURF (Speeded Up Robust Features) is a robust and useful feature detector for various vision-based applications but lacks the ability to detect symmetrical objects. This paper proposes a new symmetrical SURF descriptor to enrich the power of SURF to detect all possible symmetrical matching pairs through a mirroring transformation. A vehicle make-and-model recognition (MMR) application is then adopted to prove the practicability and feasibility of the method. To detect vehicles from the road, the proposed symmetrical descriptor is first applied to determine the ROI of each vehicle from the road without using any motion features. This scheme provides two advantages; there is no need of background subtraction and it is extremely efficient for real-time applications. Two MMR challenges, i.e., multiplicity and ambiguity problems, are then addressed. The multiplicity problem stems from one vehicle model often having different model shapes on the road. The ambiguity problem results from vehicles from different companies often sharing similar shapes. To address these two problems, a grid division scheme is proposed to separate a vehicle into several grids; different weak classifiers that are trained on these grids are then integrated to build a strong ensemble classifier. Because of the rich representation power of the grid-based method and the high accuracy of vehicle detection, the ensemble classifier can accurately recognize each vehicle.

This paper presents an approach to detect moving and static objects occurring in a video by a novel model-based tracking. The method exploits the spatial and motion coherence of objects across image frames that results from the known bounded shape distortion and object's velocity between two consecutive frames. The interframe transformation space is thus reduced to a reasonable small space of only tens or hundreds of possible states. Considering each state as a class, the object tracking process to locate objects across frames can be implemented by a classification framework, comprising a Hough-voting framework and a class-specific implicit video object model. Given a frame of the input video clip, we divide each frame of a test video clip into multiple patches which search similar model patches in the learnt implicit video object model to locate the target objects from the frames. Patch similarity is defined with respect to appearance and motion features of patches. Results show that the proposed method gives good performance on several publicly available datasets in terms of detection accuracy.

In this paper, the self organization properties of genetic algorithms are employed to tackle the problem of feature selection and extraction in ultrasound images, which can facilitate early disease detection and diagnosis. Accurately identifying the aberrant features at a particular location of clinical ultrasound images is important to find the possibly damaged tissues. Unfortunately, it is difficult to exactly detect the regions of interest (ROIs) from relatively low quality of clinical ultrasound images. The presented evolutionary optimization algorithm presents a novel approach to building features for automatic liver cirrhosis diagnosis using a genetic algorithm. The extracted features provide several advantages over other feature extraction techniques which include: automatically construct feature set and tune their parameters, ability to integrate multiple feature sets to improve the diagnosis accuracy, and ability to find local ROIs and integrate their local features into effective global features. As compared with past approaches, we span a new way to unify the processing steps in a clinical application using the evolutionary optimization algorithms for ultrasound images. Experimental results show the effectiveness of the proposed method.

In this paper, we propose a 3D template-based human action detection base on volume pattern matching. A volume pattern is obtained by detecting the principal plane from a space-time patch using the 3D moment-preserving technique. Instead of segmentation and detailed shape representation, the objective of this research is to develop and apply computer vision methods that explore the structure of a space-time template by volume pattern matching to detect the human action from a larger video sequence. A voting scheme based on volume-based Generalized Hough transform (VBGHT) is presented to provide the 3D template matching and search method, which is robust to orientation, position, and motion. Experimental results show that the proposed method gives good performance in terms of detection accuracy and robustness.

This paper presents a novel scheme to automatically and directly detect smoking events in video. In this scheme, a color-based ratio histogram analysis is introduced to extract the visual clues from appearance interactions between lighted cigarette and its human holder. The techniques of color re-projection and Gaussian Mixture Models (GMMs) enable the tasks of cigarette segmentation and tracking over the background pixels. Then, a key problem for event analysis is the non-regular form of smoking events. Thus, we propose a self-determined mechanism to analyze this suspicious event using HHM framework. Due to the uncertainties of cigarette size and color, there is no automatic system which can well analyze human smoking events directly from videos. The proposed scheme is compatible to detect the smoking events of uncertain actions with various cigarette sizes, colors, and shapes, and has capacity to extend visual analysis to human events of similar interaction relationship. Experimental results show the effectiveness and real-time performances of our scheme in smoking event analysis.

In this paper, the self organization properties of ant colonies are employed to tackle the problem of DNA copy number analysis in array CGH data, which can reveal chromosomal aberrations in the genomic DNA. These amplifications and deletions may be crucial events in the development and progression of cancer and other diseases. Accurately identifying the recurrent aberration at a particular genome location is important to find the possibly damaged genes. Unfortunately, it is difficult to exactly detect the boundaries of aberration from array CGH data with low signal-to-noise ratio. The presented ant colony optimization algorithm represents the problem as a directed graph such that the objective of the original problem becomes to find the shortest path on the graph under the problem-specific constraints. A number of artificial ants are distributed on the graph and communicate with one another through the pheromone trails which are a form of the long-term memory guiding the future exploration of the graph. The important properties of the proposed method are thoroughly investigated. The performance of the proposed method as compared to those of the state-of-the-art methods is very promising.

In this paper, we describe a novel video shot retrieval where each shot is separated into multiple video cubes. Therefore, every video shot can be represented by a linear combination of video cubes, which are used to calculate the similarity measurements among video shots in terms of video cube similarity. The position of a voxel in a cube is characterized with (x, y, t) 3D coordinates and the spatial-temporal features within video cubes are extracted with a set of analytical formulas derived from the proposed 3D moment-preserving technique. Then, the content of a video cube is approximated by three blocks generated from projecting the cube onto xy, yt and tx planes. Based on the visual patterns of xy, yt, and tx blocks, a fast video shot retrieval scheme is proposed. As compared with other key-frame based representations, the proposed cube-based video retrieval improves the retrieval accuracy without sacrificing the execution speed. Experimental results show the efficiency and effectiveness of the proposed video retrieval.

This paper proposes a parallel solution to retrieve images from distributed data sources using perceptual grouping of block-based visual patterns. The method of grouping visual patterns into image model based on generalized Hough transform is one of the most powerful techniques for image analysis. However, real-time applications of this method have been prohibited due to the computational intensity in similarity searching from a large centralized image collection. A query object is decomposed into non-overlapped blocks, where each of them is represented as a visual pattern obtained by detecting the line edge from the block using the moment-preserving edge detector. A voting scheme based on generalized Hough transform is proposed to provide object search method, which is invariant to the translation, rotation, scaling of image data. In this work, we describe a heterogeneous cluster-oriented CBIR implementation. First, the workload to perform an object search is analyzed, and then, a new load balancing algorithm for the CBIR system is presented. Simulation results show that the proposed method gives good performance and spans a new way to design a cost-effective CBIR system.

This paper proposes an object-based image retrieval using a method based on visual pattern matching. A visual pattern is obtained by detecting the line edge from a square block using the moment-preserving edge detector. It is desirable and yet remains as a challenge for querying multimedia data by finding an object inside a target image. Given an object model, an added difficulty is that the object might be translated, rotated, and scaled inside a target image. Instead of segmentation and detailed object representation, the objective of this research is to develop and apply computer vision methods that explore the structure of an image object by visual pattern detection to retrieve images from a database. A voting scheme based on generalized Hough transform is proposed to provide object search method, which is invariant to the translation, rotation, scaling of image data. Computer simulation results show that the proposed method gives good performance in terms of retrieval accuracy and robustness.

This paper presents a new color image watermarking scheme that combines color edge detection and color quantization using the principal axis analysis on 3-dimensional color space. The watermark is hidden within the data by modifying a subset of carefully selected edge points to resist both geometric distortion and signal processing attacks. Experimental results show that the robustness of the proposed scheme to resist common attacks is good.

This paper proposes an object-based coding method for very low bit-rate channels, using a method based on motion estimation with a block-based moment-preserving edge detector. In most existing object-based coding methods, only the global motion components are transmitted. However, the global motion prediction error is large, even after motion compensation using the discrete cosine transform (DCT), when images contain rapid moving objects and noise. Furthermore, the global motion-compensating method cannot result in small prediction error if the segmented objects consist of subobjects that move along different directions. The technique proposed in this paper involves segmenting moving objects from video sequences and representing objects compactly by visual-pattern approximations of the boundary. A visual pattern is obtained by detecting the line edge from a square block using the moment-preserving edge detector. Since high computational complexity is required for motion estimation using block matching, a fast block-matching method based on the visual patterns is proposed to reduce the burden of the overall coder complexity. Computer simulation results show that the proposed method gives good performance in terms of the subjective quality, the peak signal-to-noise ratio, and the compression ratio.

This paper presents a novel error concealment method for video decoding system. The proposed algorithm adaptively combines the spatial interpolation and the temporal prediction technique based on block variance and interframe correlation, to recover the lost data. The adaptive function depends on the scene change detection, motion distance and spatial information from the nearby blocks of the previous and current frames to determine the weighting of the spatial interpolation and the temporal compensation. Simulations demonstrate that the proposed technique can achieve well subjective and quantitative results, and outperforms all the others against which are compared. Even if the scene changes in the videos, this algorithm also can efficiently recover the damaged blocks for Intra(I), Predictive(P), and Bidirectional (B) frames.

The paper presents an adaptive coding control for real-time video coding systems. Based on temporal correlations, the group-of-pictures (GOP) of a video sequence is split into one basic GOP (BGOP) and many adaptive GOPs (AGOPs) and then processed accordingly. The advantage of this method is to improve the coding efficiency, particularly solving the scene change problem. Even if the coding bit rate is over the budget, the coding scheme does not require re-encoding, hence it is especially suitable for real-time coding. Based on the adaptive algorithm, the chip is realized to process the functions of picture type decision, coding rate estimation, quantization scale decision, scene-change detection and coding mode decision in the operation. The gate count of this chip is only about 2000 and its silicon area is 2.85 mm/sup 2/ using a 0.35 /spl mu/m CMOS process.

A neural-network implementation of the moment-preserving technique, which is widely used for image processing, is proposed. The moment-preserving technique can be thought of as an information transformation method which groups the pixels of an image into classes. The variables in the so-called moment-preserving equations are determined iteratively by a recurrent neural network and a connectionist neural network which work cooperatively. Both of the networks are designed in such a way that the sum of square errors between the moments of the input image and those of the output version is minimized. The proposed neural network system is applied to automatic threshold selection. The experimental results show that the system can threshold images successfully. The performance of the method is compared with those of four other histogram-based multilevel threshold selection methods. The simulation results show that the proposed technique is at least as good as the other methods.>