This paper proposes a hybrid time series forecasting framework that combines Gaussian filtering for signal denoising with a Large Language Model (LLM) reprogrammed for temporal prediction (Time-LLM), optimized considering a structured random search across multiple agents. The Gaussian filter preprocesses raw data to suppress high-frequency noise while preserving structural trends, and the Time-LLM aligns continuous time series with discrete language tokens through cross-modal reprogramming. Evaluated on hourly turbine flow data from three Brazilian hydroelectric plants. For a horizon equal to 10 steps ahead, the proposed method achieves a root mean square error of 88.3, mean absolute error of 66.9, and mean absolute percentage error of 1.8 %, outperforming state-of-the-art benchmarks, including TiDE, NHITS, NBEATS, GRU, DeepAR, TFT, TCN, Informer, PatchTST, FEDformer, KAN, TimesNet, and standard Time-LLM.

This paper presents a Fourier-enhanced dynamic sequence-to-sequence latent graph neural network (Seq2SeqLatentGNN), a deep learning architecture for multi-node spatiotemporal forecasting in hydroelectric reservoir systems. The model integrates three key components: (i) a custom Fourier layer that analyzes global temporal patterns through frequency-domain transformations, (ii) a latent correlation graph convolutional network that infers relational structures between monitoring stations without requiring predefined adjacency matrices, and (iii) an attention-based sequence-to-sequence model that processes temporal dependencies while enabling multi-step forecasting. The architecture simultaneously learns graph structure and forecasting tasks, adapting to changing spatial relationships between reservoir nodes. The proposed architecture was evaluated using a comprehensive dataset derived from 19 interconnected hydroelectric reservoirs located in southern Brazil. The dataset encompasses multiple years of high-resolution (hourly) measurements, including reservoir water levels, inflow and outflow rates, precipitation records, and energy production metrics. Experimental results demonstrate that Seq2SeqLatentGNN achieves superior performance compared to conventional statistical models and contemporary machine learning methods, as measured by standard error metrics. Analysis of the learned latent correlations reveals meaningful spatial dependencies that align with hydrological principles. The model exhibits consistent performance across varying temporal patterns, adapts to regime transitions, and captures both periodic and nonstationary dynamics. The proposed architecture contributes to spatiotemporal forecasting by combining spectral processing, dynamic graph learning, and sequence modeling in a unified framework applicable to systems with evolving connectivity patterns. • Fourier layer captures global temporal patterns in reservoir systems. • Latent graph convolution uncovers dynamic spatial relationships. • Attention model processes temporal dependencies for multi-step forecasts. • Seq2SeqLatentGNN outperforms ML models on hydroelectric data.

Efficient reasoning about the spatial and temporal structure of the environment is crucial for perception in autonomous driving , particularly in an end-to-end approach. Although different sensor modalities are employed to capture the complex nature of the environment, they each have their limitations. For example, frame-based RGB cameras are susceptible to variations in illumination conditions . However, these limitations at the sensor level can be addressed by complementing them with sensor fusion techniques, enabling the learning of efficient feature representations for end-to-end autonomous perception. In this study, we address the end-to-end perception problem by fusing a frame-based RGB camera with an event camera to improve the learned representation for predicting lateral control. To achieve this, we propose a convolutional encoder–decoder architecture called DRFuser. DRFuser encodes the features from both sensor modalities and leverages self-attention to fuse the frame-based RGB and event camera features in the encoder part. The decoder component unrolls the learned features to predict lateral control, specifically in the form of a steering angle . We extensively evaluate the proposed method on three datasets: our collected Dataset, Davis Driving dataset, and the EventScape dataset for simulation. The results demonstrate the generalization capability of our method on both real-world and simulated datasets. We observe qualitative and quantitative improvements in the performance of the proposed method for predicting lateral control by incorporating the event camera in fusion with the frame-based RGB camera. Notably, our method outperforms state-of-the-art techniques on the Davis Driving Dataset, achieving a 5.6% improvement in the root mean square error (RMSE) score.

We present adversarial event prediction (AEP), a novel approach to detecting abnormal events through an event prediction setting. Given normal event samples, AEP derives the prediction model, which can discover the correlation between the present and future of events in the training step. In obtaining the prediction model, we propose adversarial learning for the past and future of events. The proposed adversarial learning enforces AEP to learn the representation for predicting future events and restricts the representation learning for the past of events. By exploiting the proposed adversarial learning, AEP can produce the discriminative model to detect an anomaly of events without complementary information, such as optical flow and explicit abnormal event samples in the training step. We demonstrate the efficiency of AEP for detecting anomalies of events using the UCSD-Ped, CUHK Avenue, Subway, and UCF-Crime data sets. Experiments include the performance analysis depending on hyperparameter settings and the comparison with existing state-of-the-art methods. The experimental results show that the proposed adversarial learning can assist in deriving a better model for normal events on AEP, and AEP trained by the proposed adversarial learning can surpass the existing state-of-the-art methods.

Multi‐target tracking in a non‐overlapping camera network is an active research field, and one of the important problems in it is the person re‐identification problem. In this study, the authors propose an approach to improve the performance of the backbone model in the person re‐identification. Their approach focuses on training a fusion model with a shallow model and making hard triplets with relationship matrices quickly and efficiently. The proposed approach is simple, but it improves the performance of the backbone. In addition, the hard triplet mining in their process is much faster than the conventional approach. Experimental evaluation shows that the proposed approach can improve the performances of the backbone model. The proposed approach improves rank‐1 and mean average precision (mAP) performance by more than 12.54 and 15.44%, respectively, over the backbone models in the Market1501 and DukeMTMC‐reID dataset. The approach also achieves competitive performances compared with state‐of‐the‐art approaches.

In this paper, a fully automatic method for gridding bright field images of bead-based microarrays is proposed. There have been numerous techniques developed for gridding fluorescence images of traditional spotted microarrays but to our best knowledge, no algorithm has yet been developed for gridding bright field images of bead-based microarrays. The proposed gridding method is designed for automatic quality control during fabrication and assembly of bead-based microarrays. The method begins by estimating the grid parameters using an evolutionary algorithm. This is followed by a grid-fitting step that rigidly aligns an ideal grid with the image. Finally, a grid refinement step deforms the ideal grid to better fit the image. The grid fitting and refinement are performed locally and the final grid is a nonlinear (piecewise affine) grid. To deal with extreme corruptions in the image, the initial grid parameter estimation and grid-fitting steps employ robust search techniques. The proposed method does not have any free parameters that need tuning. The method is capable of identifying the grid structure even in the presence of extreme amounts of artifacts and distortions. Evaluation results on a variety of images are presented.

Cloud computing is becoming increasingly prevalent in recent years. It introduces an efficient way to achieve management flexibility and economic savings for distributed applications. To take advantage of computing and storage resources offered by cloud service providers, data owners must outsource their data onto public cloud servers which are not within their trusted domains. Therefore, the data security and privacy become a big concern. To prevent information disclosure, sensitive data has to be encrypted before uploading onto the cloud servers. This makes plain text keyword queries impossible. As the total amount of data stored in public clouds accumulates exponentially, it is very challenging to support efficient keyword based queries and rank the matching results on encrypted data. Most current works only consider single keyword queries without appropriate ranking schemes. The multi-keyword query problem was being considered only recently. MRSE [1] is one of the first research works to define and address the problem of effective yet secure ranked multi-keyword search over encrypted cloud data. However, the keyword dictionary used in MRSE is static and must be rebuilt when the number of keywords in the dictionary increases. It also has severe out-of-order problems in the matching results and does not take the keyword access frequencies into account, which greatly affects its usability. In this paper, we propose a novel approach, called MKQE, to address these issues. Only minor changes in the dictionary structure have to be done when extra keywords are introduced. We also introduce new trapdoor generation and scoring algorithms to make in-order query results. Furthermore, the keyword access frequency is considered so as to select an adequate matching file set. We conduct extensive simulations and the results prove that our approach performs much better than previous solutions.

We describe the prototype of an image retrieval system used for mobile guide applications. It allows users to send image queries of urban scenes using camera phones. The system uses a database of views of scenes to determine the poses of query views provided by the users. Information is then mapped onto the query views. Our approach is based on a fully affine invariant descriptor, coined PRIUS, which can identify corresponding building facades across widely separated views despite the highly repetitive nature of the man-made environments. The system is substantiated by experiments that show that our approach outperforms the state-of-the-art approaches such as SIFT (scale-invariant feature transform) and MSER (maximally stable external regions) in image retrieval. © 2012 Wiley Periodicals, Inc.

Research done in two aspects of robot localization and mapping are presented. An online fast learning FLANN, capable of learning location specific spatio-temporally stable visual features is developed. A technique for building oriented place maps using vestibular sensory information that can store multiple pose information of objects is investigated. Unlike most localization and mapping techniques ours does not require any depth estimation and can also handle dynamically changing environments. The system is tested in indoor environments, ranging from very simple to extremely cluttered ones. Preliminary research results show good generalization and learning capabilities of the network and improved localization using multiple oriented place maps.

This paper presents an approach for reliable matching of feature points in images of building facades with wide viewpoint changes. These images of man-made facades can render state-of-the-art techniques such as SIFT ineffective due to low texture and inclusion of repetitive items. The proposed approach integrates both appearance-based measurements and geometric measurements to find correct feature matches. The reliabilities of feature matches are reinforced with a belief network based on the assumption that if a match is a true match, it should have similarity in geometry with the other matches. The geometric measurements used here are the distances and angles between the imaginary lines connecting the pairs of matches. The most reliable match which we call the reference match is selected based on the achievement of maximum votes from other matches. The number of correct matches is expanded by selecting the matches which have similar geometry with respect to the reference match. A homography model is then applied to eliminate the wrongly selected matches. Extensive experimental results demonstrate higher reliability of our approach than state-of-the-art techniques such as SIFT, and SIFT in combination with RANSAC.

In the literature, we have seen a boom in wide-baseline matching approaches proposed for locating correspondences between images. However, wrong correspondences or the so-called outliers are still rather inevitable, especially in urban environments with the presence of repetitive structures, and/or a large dissimilarity in viewpoints. In this paper, we propose a verification model to optimize the image matching results by significantly reducing the number of outliers. Several geometric and appearance-based measurements are exploited, and conditional probability is used to compute the probability of each true correspondence. The model is validated by extensive experiments on images from the ZuBud database, which are taken in different weather conditions, seasons, and with different cameras. It is also demonstrated on a real-time application of an image-based navigation system. © 2010 Wiley Periodicals, Inc.

This paper presents a novel approach to finding point correspondences between images of building facades with wide viewpoint variations, and at the same time returning a large list of true matches between the images. Such images comprise repetitive and symmetric patterns, which render popular algorithms e.g., SIFT to be ineffective. Feature descriptors such as SIFT that are based on region patches are also unstable under large viewing angle variations. In this paper, we integrate both the appearance and geometric properties of an image to find unique matches. First we extract hypotheses of building facades based on a robust line fitting algorithm. Each hypothesis is defined by a planar convex quadrilateral in the image, which we call a “q-region”, and the four corners of each q-region provide the inputs from which a projective transformation model is derived. Next, a set of interest points are extracted from the images and are used to evaluate the correctness of the transformation model. The transformation model with the largest set of matched interest points is selected as the correct model, and this model also returns the best pair of corresponding q-regions and the most number of point correspondences in the two images. Extensive experimental results demonstrate the robustness of our approach in which we achieve a tenfold increase in true matches when compared to state of the art techniques such as SIFT and MSER.

In this paper, we consider a multi-hop network where a mobile node can operate as a host as well as a router, helping other mobile nodes to connect to the access point. We propose a network reconfiguration algorithm to re-establish connections between mobile nodes, to achieve load-balancing for wire

Our paper presents a system for efficient recognition of landmarks taken from camera phones. Information such as tutorial rooms within the captured landmarks is returned to user within seconds. The system uses a database of multiple viewpoint's images for matching. Various navigational aids and sensors are used to optimize accuracy and retrieval time by providing complementary information about relative position and viewpoint of each query image. This makes our system less sensitive to orientation, scale and perspective distortion. Multi-scale approach and a reliability score model are proposed in this application. Our system is validated by several experiments in the campus, with images taken from different resolution's camera phones, positions and times of day.

In this paper, we consider a multihop wireless network with infrastructure support (MWNI) which can be built upon existing IEEE 802.11 based WLANs to extend the coverage and provide more flexibility to adapt to the dynamics of the network. To support integrated services in MWNI, a QoS framework is proposed to achieve the following objectives: (1) allow more real-time flows to be admitted into the network, (2) optimize the performance of best-effort traffic without losing fairness and (3) enhance QoS performance of admitted real-time traffic. First, an optimized algorithm is proposed to estimate the bottleneck bandwidth. Second, a new QoS routing protocol is proposed that discovers global optimized topology based on the bandwidth estimation. Third, a two-level hierarchical scheduling algorithm is implemented at the APs. The efficiency of individual component, as well as the QoS framework is evaluated through simulation studies, and the results show our proposed QoS framework is able to enhance the QoS of real-time traffic and still achieve high network utilization.

The cooperation and coordination among the different robots and their self-governance in a workplace have become important issues in engineering design Robot soccer systems have been used as a platform to demonstrate the various researches in technology and implementation of a multi-robot cooperative environment. In this paper, a strategy-driven cooperative framework among multi-robot is proposed and investigated. Task completion is achieved through three main components: (1) sensing and representation of the dynamic environment information as the primary input parameters, (2) choice of a suitable strategy and allocation of tasks to achieve individual and inter-dependent behaviours, and (3) motion control and path planning to guide the robots to the required destination and perform the required action

With the rapid development and wide deployment of wireless local area networks (WLANs), WLAN-based positioning system employing signal-strength-based technique has become an attractive solution for location estimation in indoor environment. In recent years, a number of such systems has been presented, and most of the systems use the common nearest neighbor in signal space (NNSS) algorithm. In this paper, we propose an enhancement to the NNSS algorithm. We analyze the enhancement to show its effectiveness. The performance of the enhanced NNSS algorithm is evaluated with different values of the parameters. Based on the performance evaluation and analysis, we recommend some guidelines on optimizing the parameters of our proposed enhanced algorithm.

A number of mobile smart devices such as PDAs and sophisticated mobile phones are already available, and the trend is set to continue with more devices becoming wirelessly connectable and communicable. This implies that Wireless Personal Area Networks (WPANs) are likely to become commonplace in the near future. Since a WPAN is personal and likely to be used by one user at a time, it makes sense and would also be desirable for it to be automatically configured or customised for the current user. Conventional methods of storing and activating user settings are usually designed for configuring a single workstation and thus cannot be applied to a WPAN, a network of devices that may or may not be computers. Additionally, an inherent nature of a WPAN is the ad-hoc connectivity of its devices, which conventional methods of configuration were not designed to address. In this project, mobile agent technology was used to solve the configuration problem. A study was done on mobile agent technologies, and the Aglet system was selected for use. Research was also done on wireless networking technologies suitable for WPAN, and Bluetooth was used to connect up a simple WPAN in this project. The system worked well and was able to configure the remote PC to the desired settings in about a minute or less.

Most existing face detection approaches have assumptions, which make them applicable only under some specific conditions. In this research, we propose a combination of the color-based approach and the feature-based approach to detect faces in a color image. This combined approach will allow detecting faces in more general circumstances. As an initial step, the skin color will be utilized to separate the faces from other complex background. We develop a color-based face detector, which is used to find the face boundaries in the image. The color-based detector reliably isolates any skin-colored region found in the image. However, the regions identified may sometimes consist of other skin-colored objects as well. The facial features, e.g. eyes and mouth, will be used to verify the existence of a face within a skin. The features will be extracted from the skin region and grouped into face candidates. To reliably select the correct face candidate, belief networks are developed to represent the relationships among features and to reinforce their probabilities. Some results are included to demonstrate the capability of the combined approach in detecting faces with different poses, multiple faces, and occlusion.

Human face detection has always been an important problem for face, expression and gesture recognition. Though numerous attempts have been made to detect and localize faces, these approaches have made assumptions that restrict their extension to more general cases. In this research, we propose a feature-based face detection algorithm that can be easily extended to detect faces under different scale and orientation. Feature points are detected from the image using spatial filters and grouped into face candidates using geometric and gray level constraints. A probabilistic framework is then used to evaluate the likelihood of the candidate as a face. We provide results to support the validity of the approach, and show that the algorithm can indeed cope efficiently with faces at different scale and orientation. 1 Introduction Human face recognition is a field that has important applications in our daily activities, such as for security verification, criminal identification, and...

Present approaches to human face detection have made several assumptions that restrict their ability to be extended to general imaging conditions. We identify that the key factor in a generic and robust system is that of exploiting a large amount of evidence, related and reinforced by model knowledge through a probabilistic framework. In this paper, we propose a face detection framework that groups image features into meaningful entities-using perceptual organization, assigns probabilities to each of them, and reinforce there probabilities using Bayesian reasoning techniques. True hypotheses of faces will be reinforced to a high probability. The detection of faces under scale, orientation and viewpoint variations will be examined in a subsequent paper.

Many current human face detection algorithms make implicit assumptions about the scale, orientation or viewpoint of faces in an image and exploit these constraints to detect and localize faces. The algorithm may be robust for the assumed conditions but it becomes very difficult to extend the results to general imaging conditions. In an earlier paper (Yow and Cipolla, 1996) we proposed a feature-based face detection algorithm to detect faces in a complex background. In this paper we examine its ability to detect faces under different scale, orientation and viewpoint. The results show that the algorithm can indeed cope with a good range of scale, orientation and viewpoint variations that is typical of a subject sitting in front of a computer terminal.

This paper describes a method to detect and locate human faces in an image given no prior information about the size, orientation, and viewpoint of the faces in the image. This method uses a family of Gaussian derivative filters to search and extract human facial features from the image and then group them together into a set of partial faces using their geometric relationship. A belief network is then constructed for each possible face candidate and the belief values updated by evidences propagating through the network. Different instances of detected faces are then compared using their belief values and improbable face candidates discarded. The algorithm is tested on different instances of faces with varying sizes, orientation and viewpoint and the results indicate a 91% success rate in detection under viewpoint variation.