Abstract

This study investigates the landscape of environmental litigation in India, analyzing how environmental disputes unfold within the judiciary. It aims to identify significant patterns in case subject matter, types of appellants, and regional distribution of such cases. By examining the broader socio-legal context, the research explores how various stakeholders interact with the legal system to address environmental issues. The methodology involves assembling a dataset of legal cases and applying computational techniques to extract metadata and detect trends across multiple dimensions. Statistical analysis was used to assess correlations between case topics and appellant categories, while spatial data provided insights into regional variations in environmental disputes. Additionally, a qualitative review of relevant legislation and policies cited in key rulings helped contextualize the legal reasoning behind environmental judgments. Through a computational lens, this study uncovers the institutional dynamics that shape environmental justice in India, emphasizing the value of data-driven approaches in understanding the interplay between law, governance, and civic engagement in environmental policymaking.

Abstract

We present TabGuard, a privacy-preserving framework for an end-to-end secure Table Structure Recognition. Tab-Guard masks all the contents of the table locally and utilizes the masked table image for structure recognition. Our method is simple yet effective for detecting table cells while preserving the inherent table alignment characteristics to reconstruct tables. Our approach benefits from inductive bias, expressed through an approximated table grid which helps alleviate challenges in the detection of cells that are small or have extreme aspect ratios. Experimental results demonstrate that our solution not only establishes a new state-of-the-art on several benchmark datasets but also effectively addresses long-standing challenges associated with dense tables having complex layouts.

Abstract

The demand for contactless heartbeat monitoring systems is rapidly increasing, particularly in healthcare settings where minimizing physical contact is crucial. Conventional methods often require subjects to remain stationary to ensure accurate measurements, which restricts their use in dynamic or realworld scenarios. In this paper, we present a novel system that utilizes Texas Instruments’ 77 GHz AWR1843BOOST FMCW radar technology to circumvent this constraint. By mitigating the effects of motion artifacts, the system enables accurate heartbeat detection even when the subject is in motion. Implemented on the PYNQ-Z2 system-on-chip (SOC) platform, the system achieves an accuracy of 82.78%, providing a more robust and portable embedded solution for real-time, non-invasive heart monitoring in a dynamic environments.

Abstract

In this paper, we propose an FPGA based system for non-contact monitoring of respiration rates (RR) using a mmWave frequency modulated continuous wave (FMCW) radar. The novel FPGA based hardware architecture presented in this paper classifies normal and abnormal RR with a Support Vector Machine (SVM) model using mmWave radar-based sensing. SVM with various kernels are evaluated in order to identify the most effective approach for respiratory rate (RR) classification. SVM with a quadratic polynomial kernel shows a high classification accuracy of 95%. While the accuracy was comparable to other kernels, the quadratic polynomial kernel achieved a substantial reduction in the number of support vectors and other parameters. This optimisation results in lower resource utilisation and power consumption, making it highly efficient for FPGA implementa- tion.

Abstract

This letter presents the development of a real-time object detection system using frequency modulated continuous wave millimeter-wave (mmWave) radar signals and the python productivity for zynq ultrascale + mpsocs (PYNQ-ZU) field-programmable gate array (FPGA) board, which is widely used in advanced driving assistance system and robotic applications. A hardware FPGA platform serves as a valid embedded architecture for the purpose of validating object detection and recognition applications. We used our experiment’s point cloud images to apply different machine learning models to detect these objects. Using a top-view (TV) filter to convert 3-D point cloud images into 2-D representations made object detection more accurate. Following the use of filtration techniques, we extracted features from the filtered 2-D image using the visual geometry group (VGG) 16 model. We then assessed four machine learning models for object detection and found that the support vector machine (SVM) model and logistic regression (LR) had better results, obtaining an accuracy of 97%. Our proposed work uses mmWave radar, TV filter, VGG 16 Model, and LR to highly increase object detection accuracy over existing methods.

Abstract

Wearable heating pads are widely used as therapeutic warmers for tending to multiple health conditions. Precise temperature control is the key to ensuring the safety and effectiveness of the device due to its near-skin application. The paper presents a comparative analysis of temperature control using Proportional-Integral (PI) and Proportional-Integral-Derivative(PID) controllers based on time domain response parameters. The system utilizes an Arduino microcontroller to perform the error calculation between the set point and real-time temperature and accordingly generate PWM pulses to control the actuator driving the heating pad. The device features three different modes of operation based on the analog output voltage from an external driving circuit. Alongside, characterization data, system design and implementation of controllers, a comparative time domain response parameter analysis based on experimental data is also presented.

Abstract

The paper describes a platform for quick characterization of photodiodes exploring low reverse bias performance for various sensing applications. We demonstrated the method on a widely used photodiode, BPW34. An Arduino-based automated system is utilized to sweep biasing voltages and record photocurrent responses using a sense resistor. The BPW34 photodiode shows a higher response for red LASER light, with a photocurrent of around 260 μA. Various light sources are also tested and the results are reported. These results are analyzed and interpreted with context to the existing literature.

Abstract

In this paper, we present an approach for integrated task and motion planning based on an AND/OR graph network, which is used to represent task-level states and actions, and we leverage it to implement different classes of task and motion planning problems (TAMP). Several problems that fall under task and motion planning do not have a predetermined number of sub-tasks to achieve a goal. For example, while retrieving a target object from a cluttered workspace, in principle the number of object re-arrangements required to finally grasp it cannot be known ahead of time. To address this challenge, and in contrast to traditional planners, also those based on AND/OR graphs, we grow the AND/OR graph at run-time by progressively adding sub-graphs until grasping the target object becomes feasible, which yields a network of AND/OR graphs. The approach is extended to enable multi-robot task and motion planning, and (i) it allows us to perform task allocation while coordinating the activity of a given number of robots, and (ii) can handle multi-robot tasks involving an a priori unknown number of sub-tasks. The approach is evaluated and validated both in simulation and with a real dual-arm robot manipulator, that is, Baxter from Rethink Robotics. In particular, for the single-robot task and motion planning, we validated our approach in three different TAMP domains. Furthermore, we also use three different robots for simulation, namely, Baxter, Franka Emika Panda manipulators, and a PR2 robot. Experiments show that our approach can be readily scaled to scenarios with many objects and robots, and is capable of handling different classes of TAMP problems.

Abstract

Automatic syllable stress detection is an essential component in Computer assisted language learning (CALL) systems to guide non-native language learners. In English, each word typically contains only one primary stressed syllable. However, standard loss functions, such as Binary Cross-Entropy (BCE), often result in predictions where multiple syllables may be stressed or none at all. As a result, automatic syllable stress detection models frequently require an additional post-processing step to ensure that only one syllable is stressed per word. This reliance on post-processing suggests that the model is not fully capturing the stress patterns accurately. To address this issue, we propose an adaptive weighted loss function that builds upon the Stress Intensity Modulation Loss proposed in our recent work of Post-Net. This adaptive weighted loss function is designed to enforce the constraint of a single primary stressed syllable directly during model training. We integrate this loss function into the previously proposed Post-Net (PN_DNN) and on a new architecture which is a hybrid of Post-Net and LSTM (PN_DLSTM). Their performance is compared against the state-of-the-art models trained with standard BCE loss. Experiments conducted on the ISLE corpus reveal that both the models trained only with BCE loss show a significant accuracy gap between with and without post-processing. In contrast, when these models are trained on the proposed adaptive weighted loss function, the gap is narrowed in both the models. Between the two models, the highest reduction is observed in PN_DNN with a decrease from 3.87% to 2.45% & 4.6% to 3.75% for GER & ITA respectively. This indicates that the adaptive weighted loss function effectively captures the linguistic constraint during training, reducing the need for post-processing.

Abstract

Automatic syllable stress detection is a crucial component in Computer-Assisted Language Learning (CALL) systems for language learners. Current stress detection models are typically trained on clean speech, which may not be robust in real-world scenarios where background noise is prevalent. To address this, speech enhancement (SE) models, designed to enhance speech by removing noise, might be employed, but their impact on preserving syllable stress patterns is not well studied. This study examines how different SE models, representing discriminative and generative modeling approaches, affect syllable stress detection under noisy conditions. We assess these models by applying them to speech data with varying signal-to-noise ratios (SNRs) from 0 to 20 dB, and evaluating their effectiveness in maintaining stress patterns. Additionally, we explore different feature sets to determine which ones are most effective for capturing stress patterns amidst noise. To further understand the impact of SE models, a human-based perceptual study is conducted to compare the perceived stress patterns in SE-enhanced speech with those in clean speech, providing insights into how well these models preserve syllable stress as perceived by listeners. Experiments are performed on English speech data from non-native speakers of German and Italian. And the results reveal that the stress detection performance is robust with the generative SE models when heuristic features are used. Also, the observations from the perceptual study are consistent with the stress detection outcomes under all SE models.