Saritha Saladi

The growing deployment of autonomous Unmanned Aerial Vehicles (UAVs) in surveillance, disaster monitoring, and remote inspection applications requires reliable communication frameworks capable of operating beyond traditional ground control station (GCS) limitations. This paper presents a ground station free UAV communication architecture that enables secure global connectivity and autonomous mission continuity. The proposed framework integrates the lightweight Message Queuing Telemetry Transport (MQTT) messaging protocol with a virtual private network (VPN) overlay implemented using WireGuard through Tailscale over mobile networks. This architecture enables encrypted communication and remote telemetry monitoring among distributed UAV nodes without direct ground station dependency. In addition, a fault tolerant mechanism is introduced to detect mission interruption conditions such as low battery, communication loss, or operational failure. Upon detecting such events, the system autonomously triggers the deployment of a backup UAV to ensure uninterrupted mission execution. Unlike existing UAV communication frameworks that address networking, security, or fault tolerance independently, the proposed architecture provides an integrated approach combining decentralized communication, secure connectivity, and autonomous recovery. The system is validated through real world UAV experiments conducted over mobile networks, demonstrating reliable communication and seamless mission handover under fault conditions. The proposed framework provides a practical foundation for secure and resilient UAV operations in mission critical environments.

Brain tumor segmentation from MRI’s and PET has always been a challenging and time-consuming phase for radiologists, due to low sensitivity boundary region pixels in this image modality. Deep learning-based image segmentation is the hot research topic in recent days. Among all other deep learning models, U-Net-based variants are the most used models to segment medical images with respect to different modalities. In this paper, a Permutate version of the U-Net architecture was designed that precisely and automatically detects the boundaries of the tumour area and segments tumour regions from the fused image. There are two stages to the proposed work. In the first stage Principal component analysis (PCA) is used to fuse the MRI-PET images to enhance the fused image’s quality and improved interpretation. Later, a Permutate U-Net architecture is employed to precisely segment tumour region from the fused image. Further designed model performance is assessed using Dice Coefficient, intersection over union score (IoU) and accuracy with brain tumour segmentation challenge BraTS datasets of 2015, 2020 and 2021. Our proposed method demonstrates promising results that are superior to existing deep learning model and comparatively higher than the existing methods.

Brain Magnetic Resonance image diagnostics employs image processing, but aberrations such as Intensity Inhomogeneity (IIH) distort the image, making diagnosis difficult. Clinical diagnostic methods must address IIH discrepancies in brain MR scans, which occur often. Accurate brain MR image processing is difficult but required for clinical diagnosis. In this study, we introduced a more energy-efficient intensity inhomogeneity correction (IIC) method that makes use of the Modified Energy-based Generative Adversarial Network. This method uses reconstruction error in the discriminator architecture to save energy by altering the cost function. The generator’s performance is also improved by this reconstruction error. As the reconstruction error decreases, the discriminator collects latent information from real images to enhance output. To prevent mode collapse, the model has a drawing away term (PT). The generator design is improved by using skip connections and information modules that collect features at various scales. The suggested method beats state-of-the-art methods in metrics such as Peak Signal to Noise Ratio (PSNR), Structural Similarity Index (SSIM), Multi-Scale Structural Similarity Index (MSSSIM), Mean Squared Error (MSE), and Root Mean Square Error (RMSE).

MRI scanning has shown significant growth in the detection of brain tumors in the recent decade among various methods such as MRA, X-ray, CT, PET, SPECT, etc. Brain tumor identification requires high exactness because a minor error can be life-threatening. Brain tumor disclosure remains a challenging job in medical image processing. This paper targets to explicate a method that is more precise and accurate in brain tumor detection and focuses on tumors in neonatal brains. The infant brain varies from the adult brain in some aspects, and proper preprocessing technique proves to be fruitful to avoid miscues in results. This paper is divided into two parts: In the first half, preprocessing was accomplished using HE, CLAHE, and BPDFHE enhancement techniques. An analysis is the sequel to the above methods to check for the best method based on performance metrics, i.e., MSE, PSNR, RMSE, and AMBE. The second half deals with the segmentation process. We propose a novel ARKFCM to use for segmentation. Finally, the trends in the performance metrics (dice similarity and Jaccard similarity) as well as the segmentation results are discussed in comparison with the conventional FCM method.

In the field of medical sciences, automatic detection of tumor using magnetic resonance (MR) brain images is a major research area. The goal of the proposed work is to identify the tumors in MR images using segmentation methods and to locate the affected regions of the brain more accurately. Medical images have vast information but they are difficult to examine with lesser computational time. An innovative process is proposed to extract tumor cells using the discrete wavelet transform (DWT). After extracting features with DWT feature reduction is carried out with the principal component analysis (PCA). Modified fuzzy C means (MFCM) technique is used for segmenting the tumor cells. The efficiency of the proposed method to identify different abnormalities in real MR images for intracranial neoplasm detection, tuberculoma, and bilateral thalamic fungal granulomas identification is tested. The results obtained are shown in‐terms of Accuracy, Dice Similarity Index (DSI), and Jaccard Index (JI) measures. The performance of the proposed method is tested in terms of performance measures like Accuracy, DSI, and JI. These results are compared with the conventional fuzzy C means (FCM) method.

Diagnose the abnormalities is the hardest step in the whole medical process. Early detection of abnormalities can save lot of time, efforts and resources. The wavelet transform is one of the better methods, to achieve excellent results in terms of accuracy. So, we proposed this new method for automated classification of MRI (Magnetic resonance imaging) brain images as normal or abnormal. PPCA is used for the feature reduction process as to get decrease the computation time and complexity. SVM (support vector machine) is a classifier and to optimise the weights of the SVM, BBO (biogeography based optimization) was used. The results we obtained in terms of accuracy is superior to the previously proposed methods. Three factors accuracy.precision and sensitivity from the confusion matrix are considered into account for the evaluation of the proposed method.

Distinct algorithms were developed to segment the MRI images, to satisfy the accuracy in segmenting the regions of the brain. In this paper, we proposed a novel methodology for segmenting the MRI brain images using the clustering techniques. The Modified Fuzzy C-Means (MFCM) algorithm is pooled with the Artificial Bee Colony (ABC) algorithm after denoising images, features are extracted using Principal Component Analysis (PCA) for better results of segmentation. This improves the ability to extract the regions (cluster centres) and cells in the normal and abnormal brain MRI images. The comparative analysis of proposed methodology with existing FCM, ABC algorithms is evaluated in terms of Minkowski score. The proposed MFCM-ABC method is more robust and efficient to hostile noise in images when compared to existing FCM and ABC methods.