<!-- *** Custom HTML *** --> Deep learning techniques have become increasingly popular for classifying large-scale image and video data. Remote sensing applications require robust search engines to retrieve similar information dependent on an example-based query instead of a tag-based query. Deep features can be extracted automatically by training raw data without having any domain-specific knowledge. However, the training time for a massive amount of multimedia datasets is high. Training complexity is reduced using pre-trained GoogleNet weights for initial feature extraction. To fine-tune the feature vector and reduce the dimensionality, a one dimension convolutional neural network (1D-CNN) is applied. There is a loss of information while resizing the input image to a pre-trained network with an acceptable input size. We proposed a new feature set by integrating handcrafted features at detailed scales and deep features to improve the system’s efficiency. The curvelet transform was used to decompose the image into coarse and detailed scales. Haralick texture features were extracted from the detail coefficients in four directions and fused with fine-tuned deep features. The proposed feature set was assessed using standard performance metrics from the literature. The proposed technique achieved improved performance with 89% accuracy for retrieval of the top 50 relevant results.
<!-- *** Custom HTML *** --> Image fusion is an important technique in remote sensing to improve visual interpretation and classification. Pansharpening is the procedure of fusing panchromatic (PAN) and multispectral images to produce high spatial and spectral resolution images. Synthesized pansharpening is performed on Linear Imaging Self-Scanning Sensor III and Advanced Wide Field Sensor data, which are freely available and provided by the National Remote Sensing Center. The Adaptive Neuro-Fuzzy Inference System (ANFIS) in multiscale transform domain for multisensor image fusion application is evaluated. The state-of-the-art method has been evaluated by various quality metrics. The computational cost of ANFIS with wavelet, contourlet, shearlet, and curvelet transform is investigated. This study proves that curvelet with ANFIS-based fusion technique outperformed state-of-the-art techniques. The application will be used to incorporate the missing spectral information in the high spatial resolution PAN image to identify objects, highlighting the regions clearly.
Maternal mortality is a major public health concern worldwide. It is the number of preventable deaths that occur each year due to pregnancy and childbirth. The research investigates how machine learning may be used to minimize maternal mortality. Historical data on maternal health is used to develop predictive models, early detection systems, and resource allocation techniques. Machine learning helps to identify risk factors, monitor vital signs, and improve access to care. This allows for targeted interventions and better healthcare delivery. The challenges of data accessibility and model interpretation are addressed, highlighting the ethical and equitable applications of machine learning in maternal healthcare. This study emphasizes the potential of machine learning to reduce maternal mortality rates and the pressing need for its incorporation into healthcare systems worldwide.
This paper introduces an innovative content-based image retrieval system for precise and effective retrieval of satellite images. The system integrates liquid autoencoders with shearlet transforms to derive robust feature representations from satellite images. Empirical results demonstrate that the proposed system achieves a high retrieval accuracy of more than 95% on different datasets and is comparable with several state-of-the-art approaches. The efficiency of the system is further indicated by its fast inference time and moderate memory needs, making it suitable for large image databases. This study contributes to the literature on satellite image analysis by introducing a valid tool for several applications, including environmental monitoring and land-use mapping. The new method contributes value to the existing literature in Content-based image retrieval by showing the suitability of liquid autoencoders for satellite image retrieval.
Automated blood cell classification is crucial for hematological analysis, yet the scarcity of annotated medical datasets challenges deep learning models. This study presents a novel semi-supervised Elastic Generative Adversarial Network that enhances classification accuracy, improves synthetic image generation quality, and reduces computational complexity. The model utilizes image hallucination, which generates artificial but realistic images to supplement datasets, addressing data imbalance and improving model generalization. The generated images exhibit high fidelity, meaning they closely resemble real blood cell images in texture, morphology, and structural details, ensuring their effectiveness in training deep learning models. Compared to the conventional semi-supervised Wasserstein Generative Adversarial Network, the proposed model achieves a 5.5% increase in classification accuracy, a 34.5% reduction in computational complexity, and a 21.0% decrease in training time, demonstrating superior efficiency and scalability. In addition, it greatly enhances Fréchet Inception Distance and Wasserstein distance, which verifies improved realism and diversity of synthetic samples. Complementing pruning-based optimization ensures reduced model deployment size, rendering it appropriate for real-time medical diagnosis. Through image hallucination, the proposed method sufficiently alleviates data deficiency and improves the reliability of automated blood cell classification, qualifying it as a prospective tool for advanced hematological studies and diagnosis purposes.