PMPushpa Mamoria

Pushpa Mamoria

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Pushpa Mamoria has a total of 18 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Joint detection and state recognition of traffic lights and signs through a lightweight hybrid CNN–transformer framework

    by Hemant Kumar, Pushpa Mamoria, Deepak Kumar Dewangan

    Accurate detection of traffic lights and signs is a critical task for safe autonomous vehicle systems (AVS). Despite advances in deep learning, most methods treat traffic light detection, traffic sign detection, and state recognition as separate tasks, while CNN-based models lack global context modeling and transformer-based approaches incur high computational cost. To address these limitations, this study presents a unified and lightweight YOLOv11–EfficientViT framework that jointly performs traffic light detection, traffic sign detection, and traffic light state recognition within a single end-to-end architecture. The proposed method integrates convolutional block attention modules (CBAM) into the YOLOv11 backbone for enhanced local feature refinement and incorporates an efficient vision transformer (EfficientViT) module to capture global contextual dependencies with reduced computational cost. A multi-task loss formulation is adopted to simultaneously optimize localization and semantic state recognition. Experiments were conducted on curated, task-focused subset of BDD100K dataset for traffic light detection and state recognition and TT100K dataset for traffic sign detection. On BDD100K, the proposed model achieved 99.4% mean average precision (mAP@0.5), 92.2% mAP@[0.5:0.95], and 97.7% traffic light state recognition accuracy under the defined experimental setting while maintaining inference at 3.8 ms per frame. On TT100K, the framework attained 96.5% mAP@0.5 and 81.2% mAP@[0.5:0.95] with only 2.5M parameters. Cross-dataset evaluation on subsets of LISA and Bosch datasets further indicates promising cross-dataset transferability with limited performance variation under domain shift. The results indicate that the hybrid integration of attention-enhanced convolutional features and lightweight transformer-based global reasoning enhances detection performance and semantic consistency without compromising computational efficiency.

  • Open Access

    PCU-ST-GCN: A Spatio-Temporal Graph Convolutional Network with Cultural Event Propagation for Heterogeneous Indian Traffic Forecasting

    by Nikhil Shukla, Pushpa Mamoria

    Urban traffic management in Indian Tier-2 cities encounters a fundamental forecasting challenge: demand at any given signalised intersection is not solely driven by local conditions but by festival-induced flows that propagate along inter-city corridors over multiple days. Existing single-node time-series models and fixed-adjacency graph networks cannot represent this cross-city demand propagation. This paper introduces PCU-ST-GCN, a spatio-temporal graph convolutional network specifically designed for heterogeneous Indian traffic. Node feature vectors are constructed from five IRC:106-1990 Passenger Car Unit sub-components — two-wheeler, three-wheeler, car, light commercial vehicle, and heavy goods vehicle — encoded separately to enable vehicle-class-specific propagation learning. A Cultural Event Propagation Module (CEPM) applies learnable element-wise gating conditioned on festival severity level and a fourteen-day pre-festival countdown, encoding the gradual demand build-up characteristic of major Indian festivals such as Diwali and Kumbh Mela. An adaptive adjacency matrix A* is learned end-to-end from node embeddings, recovering functional cross-city dependencies from intersection count data alone. The model is trained and tested on 69,121 fifteen-minute observations at twelve signalised intersections across the Kanpur-Lucknow-Prayagraj National Highway 19 corridor collected from June to November 2025. PCU-ST-GCN achieves Mean Absolute Percentage Error of 7.2% at the t+15 minute horizon, representing a 25% relative improvement over standard ST-GCN, with R²=0.92. CEPM reduces Kumbh Mela prediction error at Prayagraj nodes by an average of 4.1 percentage points, with the largest gain of 5.8 percentage points at Zero Road — the primary Mela access route. Ablation study confirms that PCU decomposition, adaptive adjacency, and CEPM each contribute independently to overall accuracy improvement.

  • VEDA-Net: A Lightweight CNN Framework with Differential Attention and Edge-Aware Fusion for Real-Time Vehicle Detection

    by Hemant Kumar, Pushpa Mamoria

    Vehicle detection with a camera-based vision sensor in real time settings is critical for ensuring both the dependability and safety of AVS. In this paper we propose vehicle detection architecture using differential attention mechanisms and edge aware features (VEDA-Net). This is a lightweight CNN model which uses the Edge-Aware Feature Extraction (EAFE) technique to extract edge aware features from images. Dual Path Fusion (DPF) for fusing dual paths of information; and Differential Attention Module (DAM) to enhance the detection robustness in complex traffic environment. Additionally, VEDA-Net has a Compact Head for Detection (CHD) module that outputs bounding box and class probability through an anchor free regression method. Due to its low latency characteristics, VEDA-Net is suitable for use as an edge-based Autonomous vehicle system (AVS) perception module. On the COCO traffic dataset, VEDA-Net achieved an accuracy rate of 81%, a recall rate of 79%; and mAP@0.5 of 94%. Compared to YOLOV5s; Efficient Det-D0; and Tiny-YOLOv4, VEDA-Net's performance increased by 12%; 6%; and 16% respectively. Averaging over 500 epochs at a frame rate of 30 FPS, the average inference latency was 21.28ms. Based on qualitative results, the model demonstrated high levels of robustness under various levels of occlusions; different lighting conditions; and real-world clutter. By focusing on structural feature extraction, and selectively attending to consistent spatial regions in vehicles, VEDA-Net provides high detection robustness. As such, due to its modular and compact nature, VEDA-Net will be suitable for use as a detection module within edge-based AVS applications that require rapid, accurate, and light-weight perception modules.

  • Open Access

    Deep Learning–Based STA Channel Estimation Enhancement Using Optimized Training Algorithm in Vehicular Communications

    by Ekram Ahmed Tuha, Rajeev Kumar Shakya, Pushpa Mamoria, Demissie Jobir Gelmecha

    Wireless communication plays a pivotal role in intelligent transportation systems, with vehicular communication being a crucial application, particularly with the IEEE 802.11p standard. However, the time‐varying nature of vehicular communication channels, characterized by signal propagation in multiple directions and Doppler frequency shifts, poses challenges, resulting in a decline in signal strength. Reliable and accurate channel estimation is crucial due to high mobility and time‐varying channel characteristics. To address these issues, the STA‐LRDNN channel estimation technique is proposed to improve the estimation in diverse channel conditions through efficient training of channel estimates by compensating for errors. Proper selection of DNN hyperparameters is obtained in the proposed channel estimation. The proposed approach is derived from the traditional spectral time averaging channel estimator to further enhance performance, learning channel statistics more effectively for high‐mobility vehicular scenarios. The proposed technique has been compared with existing methods using the bit error rate (BER) and the normalized mean squared error (NMSE) performance, considering the different vehicular channel models. The proposed STA‐LRDNN shows its robustness for high‐mobility vehicular scenarios. In addition, it is observed that the STA‐LRDNN achieves a 20 dB SNR gain for 10 −2 BER and as low as 10 −5 BER at 16 QAM. For high mobility cases, it is 10 −3 BER at 16 QAM with SNR of 30 dB. Notably, it exhibits remarkable results in low mobility scenarios as well due to the adaptive nature of DNNs.

  • Urban Computing Solutions in Healthcare Edge Computing

    by Basudeo Singh Roohani, Nitin Sharma, Vinay Kumar Kasula, Pushpa Mamoria, Nirav Narendrakumar Modh, Alok Kumar, Vineeta Singh

    Healthcare is on the verge of revolutionary change through the combination of urban computing solutions with edge computing in the period of urban expansion and technological progress. The key objectives of the chapter will focus on understanding how edge computing can optimize data processing at the source, minimizing latency and enabling immediate decision-making in urgent healthcare scenarios, and identifying specific challenges faced by urban healthcare systems, including data overload, resource allocation inefficiencies, and accessibility issues in densely populated areas. The chapter uses a combination of qualitative and quantitative research methods as its methodology framework. A study of urban healthcare facilities through different case examples evaluates positive results as well as best practices in implementation. The book chapter investigates the fundamental framework of edge computing within healthcare systems which specifically targets main goals to improve patient outcomes together with operational refinement and immediate data evaluation capabilities.


  • Video Steganography Techniques: A Comprehensive Review and Performance Evaluation

    by Hemant Kumar, Pushpa Mamoria, Shivani Kumari, Manoj Kumar Misra