In surveillance environments, detecting anomalies requires understanding the contextual dynamics of the environment, human behaviors, and movements within a scene. Effective anomaly detection must address both the where and what of events, but existing approaches such as unimodal action-based methods or LLM-integrated multimodal frameworks have limitations. These methods either rely on implicit scene information, making it difficult to localize where anomalies occur, or fail to adapt to surveillance specific challenges such as view changes, subtle actions, low light conditions, and crowded scenes. As a result, these challenges hinder accurate detection of what occurs. To overcome these limitations, our system takes advantage of features from a lightweight scene classification model to discern where an event occurs, acquiring explicit location-based context. To identify what events occur, it focuses on atomic actions, which remain underexplored in this field and are better suited to interpreting intricate abnormal behaviors than conventional abstract action features. To achieve robust anomaly detection, the proposed Temporal-Semantic Relationship Network (TSRN) models spatio-temporal relationships among multimodal features and employs a Segment-selective Focal Margin loss (SFML) to effectively address class imbalance, outperforming conventional MIL-based methods. Experimental results on public datasets demonstrate that the proposed system effectively reduces false alarms while maintaining robustness and practicality for real-world surveillance applications.

Segmenting the retinal layers in optical coherence tomography (OCT) images helps to quantify the layer information in early diagnosis of retinal diseases, which are the main cause of permanent blindness. Thus, the segmentation process plays a critical role in preventing vision impairment. However, because there is a lack of practical automated techniques, expert ophthalmologists still have to manually segment the retinal layers. In this study, we propose an automated segmentation method for OCT images based on a feature-learning regression network without human bias. The proposed deep neural network regression takes the intensity, gradient, and adaptive normalized intensity score (ANIS) of an image segment as features for learning, and then predicts the corresponding retinal boundary pixel. Reformulating the segmentation as a regression problem obviates the need for a huge dataset and reduces the complexity significantly, as shown in the analysis of computational complexity given here. In addition, assisted by ANIS, the method operates robustly on OCT images containing intensity variances, low-contrast regions, speckle noise, and blood vessels, yet remains accurate and time-efficient. In evaluation of the method conducted using 114 images, the processing time was approximately 10.596 s per image for identifying eight boundaries, and the training phase for each boundary line took only 30 s. Further, the Dice similarity coefficient used for assessing accuracy gave a computed value of approximately 0.966. The absolute pixel distance of manual and automatic segmentation using the proposed scheme was 0.612, which is less than a one-pixel difference, on average.

This paper introduces a novel, model-based chemical exchange saturation transfer (CEST) magnetic resonance imaging (MRI), in which asymmetric spectra of interest are directly estimated from complete or incomplete measurements by incorporating subspace-based spectral signal decomposition into the measurement model of CEST MRI for a robust z-spectrum analysis. Spectral signals are decomposed into symmetric and asymmetric components. The symmetric component, which varies smoothly, is delineated by the linear superposition of a finite set of vectors in a basis trained from the simulated (Lorentzian) signal vectors augmented with data-driven signal vectors, while the asymmetric component is to be inherently lower than or equal to zero due to saturation transfer phenomena. Spectral decomposition is performed directly on the measured spectral data by solving a constrained optimization problem that employs the linearized spectral decomposition model for the symmetric component and the weighted Frobenius norm regularization for the asymmetric component while utilizing additional spatial sparsity and low-rank priors. The simulations and in vivo experiments were performed to demonstrate the feasibility of the proposed method as a reliable molecular MRI.

Optical methods are becoming powerful clinical tools for non-invasive diagnosis and brain-computer interfacing. The recent development of near infrared spectroscopy, optical coherence tomography, and photoacoustic imaging enable quantitative mapping of brain activity and microscale histological imaging. In this work, we presented principles and applications of these optical methods.

This paper introduces the restoration method of degraded images distorted by inherent structure of fiber bundle. The criterion considered for the restoration is based on morphological processing and notch filter, removing the particular frequency, with auto peak detection. The proposed algorithm achieved effective removal of pixelated artifact such as regular honeycomb pattern by finding particular peaks in Fourier domain. As a series of procedure, we have found optimized parameter for decoupling the noise. Experimental results of contrast-to-noise ratio (CNR) demonstrate numerical enhancement from 3.9dB to 16.7dB. Our proposed method can also be expected to apply other periodic pattern noise.

Optical code division multiple access (OCDMA) has been extensively studied for the evaluation of system performance. In the phase modulated OCDMA system, the coherent balance detector receiver was used with the phase-locked loop (PLL) for code and carrier phase synchronization. Due to the implementation limitations of phase coded OCDMA and the inefficiencies for noncoherent detection, this paper considers the homodyne correlation receiver using dual balance detector for the estimation of transmission performance. A general model applicable to any number of users is set up, including noises such as shot noise and equivalent receiver circuit noise in the receiver model as well as multiuser interference. Moreover, the code correlation properties are observed for the chosen prime code. The influence of the code phase alignment condition in its synchronizing process on system performance is also analyzed. Rather than concentrating the study on the strategy for the code synchronization procedure followed by the computation of the average acquisition time as the performance of interest, the probabilities of false alarm and bit error depending on the state of code phase alignment are on focus.