Soil loss due to water erosion poses a significant threat to agricultural sustainability and food security, particularly across semi-arid Mediterranean regions. In Tunisia, this phenomenon is intensified by the interplay of soil properties and spatiotemporal dynamics of Land Use and Land Cover (LULC). Accurately assessing soil erosion through physically based, distributed, and continuous models is critical for effective land management. However, such assessments demand high-resolution, temporally dynamic input data, making manual parameterization both complex and time-consuming. This study aims to enhance the parameterization of the ANSWERS-2000 hydrological model by leveraging Google Earth Engine (GEE) for the automated extraction of key MODIS-derived input, including annual Land cover, phenological transition dates (green-up, maturity, senescence, and dormancy), land Surface Temperature, and solar radiation, over the M’richet El Anse catchment (1.65 km2, Siliana, Tunisia), spanning January 1, 2001, to August 31, 2002. GEE extracted inputs replaced traditionally estimated parameters for crop rotation scheduling and C-factor dynamics within ANSWERS-2000. Model performance was evaluated against observed sediment yield data from bathymetric surveys and hydrological records using statistical indicators, including the Nash-Sutcliffe Efficiency (NSE) and coefficient of determination (R²). Results show that GEE enhanced model achieved an R² of 0,81 and an NSE of 0.72 outperforming simulation based on traditionally estimated parameters. Spatial mapping of annual soil loss revealed a highly heterogeneous erosion pattern across the catchment, with negative values indicating sediment detachment zone and positive values indicating deposition areas. These findings demonstrate that integrating GEE based remote sensing data extraction into physically and continuous based erosion modelling significantly improves input accuracy and model predictive performance, offering a scalable and reproductible approach for water erosion assessment in data scarce semi-arid environments.

In this paper, we address the problem of children drawing classification by enhancing MobileNet through Salp Swarm Optimization (SSO). The proposed solution increases the classification accuracy with less computational processing time. The experimental setup uses the Kids’ Hand Movement Dataset consisting of 1000 pictures from fifty children with normal and fifty children with not-normal hand drawings streams. Especially, GridSearchCV-based hyperparameter is applied when the MobileNet model is tuned to avoid overfitting and reach the best performance. The experiments showed that the proposed method provided a notable performance gain in terms of accuracy in comparison with baseline model.

EEG-based biometric authentication has emerged as a secure alternative to conventional authenticationmethods, owing to its resistance to spoofing and inherent movement/image individual variability. This studyevaluated the performance of various classification models in the EEG motor movement/image dataset,which comprises 1,526 sessions recorded from 109 subjects using 64 EEG channels at a sampling rate of160 Hz. A comprehensive set of 1,600 features per session was extracted in the time, frequency, and timefrequencydomains. Following standard pre-processing and normalization, the models were trained in astratified 70/30 training test split using features standardized to zero mean and unit variance.We systematically compared traditional machine learning classifiers, ensemble methods, and deep learningarchitectures. Hyperparameter tuning was performed uniformly across all the models. The Ridge Classifierachieved the highest accuracy (93.8%), followed by Logistic Regression (91.27%) and MLP (89.96%),demonstrating the strength of linear and shallow neural models on engineered EEG features. In contrast,deep learning models, including CNN, LSTM, GRU, and BiLSTM, recorded significantly lower accuracy( 0.87%) because of limited training data and the use of pre-extracted statistical features instead of rawtime-series input, which restricted their ability to learn temporal patterns.These findings indicate that traditional machine-learning models, when applied to well-crafted features,remain highly competitive for EEG-based authentication. They offer a favorable balance between performance,computational efficiency, and interpretability, whereas deep learning approaches require furtheradaptation to the structure and scale of EEG data.

Hybrid neural network applications for cancer diagnosis mark a substantial development in medical science. The creation, integration, and potential of hybrid neural networks across different cancer types are rigorously analyzed in this paper. This review attempts to provide a thorough understanding of their function in early detection and customized management of cancer by analyzing their adaptability and efficacy, independent of imaging modalities. The review provides insights into potential future avenues for utilizing hybrid neural networks to improve the results of cancer treatment through this investigation.

Autism Spectrum Disorders (ASD) are one of the most serious health problems that our generation is facing [1]. It affects around one out of every 54 children and causes issues with social interaction, communication [2] and repetitive behaviors [3]. The development of full biomarkers for neuroimaging is a crucial step in diagnosing and tailoring medical care for autism spectrum disorder [4]. Volumetric studies focused on 3D MRI texture features have shown a high capacity for detecting abnormalities and characterizing variations caused by tissue heterogeneity. Recently, it has been the interest of comprehensive studies. However, only a few studies have aimed to investigate the link between object texture and ASD. This paper suggests a framework based on geometric texture features analyzing the variations between ASD and development control (DC) subjects. Our study uses 1114 T1-weighted MRI scans from two groups of subjects: 521 individuals with ASD and 593 controls (age range: 6–64 years) [5], divided into three broad age groups. We then computed the features from automatically labeled subcortical and cortical regions and encoded them as texture features by applying seven global Riemannian geometry descriptors and eight local features of standard Harlicks quantifier functions. Significant tests were used to identify texture volumetric differences between ASD and DC subjects. The most discriminative features are selected by applying the Correlation Matrix, and these features are used to classify the two classes using an Artificial Neural Network analysis. Preliminary results indicate that in ASD subjects, all 15 structure-derived features and subcortical regions tested have significantly different distributions from DC subjects.

In light of the abundance and heterogeneity of data stemming from diverse, highly scalable, and distributed sources, information systems encounter novel challenges. The colossal datasets, characterized by diverse types, expansive storage ca-pacities' and unprecedented communication speeds, necessitate addressing increasingly intricate queries. Query optimization emerges as a predominant challenge within the realm of big data, primarily driven by concerns related to performance and cost. Consequently, we introduce our approach to optimizing the performance of HiveQL query action plans through the utilization of genetic algorithms, which exhibit efficacy across various domains, particularly in the domain of combinatorial optimization. Genetic algorithms circumvent the elevated costs associated with optimization efforts and provide adaptability by operating independently of problem-specific knowledge. These attributes render them a viable solution for addressing the query optimization predicament. The outcomes of our study underscore significant enhancements in performance, affirming the effective-ness of the proposed genetic algorithm-based approach.

Soil loss by water erosion is a major risk affected a food security in the world and especially in the Mediterranean region. In Tunisia, this phenomenon is accentuated by soil, land use/cover characteristics. The assessment of soil loss using a physically-based, distributed, and continuous model is very important, but it is also highly requested by daily and high resolution data. These tasks are very difficult for human analysis. Artificial Intelligence and a subtype of machine learning called deep learning is used to perform with more speed, consistency and perhaps more accurate than humans can perform. This study is focused on M’richet El Anse watershed, characterized by a moderate Mediterranean climate and a high spatial and temporal heterogeneity of land use proprieties. We used the Google Earth Engine to execute a deep learning algorithm to extract, temperature, solar radiation, land use, land cover change, and rotation from MODIS image from 01/01/1994 to 31/08/2002. The outputs of ANSWERS-2000 model were evaluated by comparison between the predicted and observed values using statistical coefficients including coefficient of determination and Nash–Sutcliffe efficiency. Results show that the model outperformed the usually estimated input parameters in the assessment of the annual soil erosion (R-squared: 0.81 and NSE: 0.72). Finally, the use of deep learning for big spatial data makes the use of the hydrological and water erosion model, ANSWERS-2000, more and best adopted for the estimation, quantification, and spatial variation of water erosion at the watershed scale.

This paper introduces a novel steganography method for embedding and extracting a secret message from an image file using three stages. In the first stage, Binary Robust Invariant Scalable Keypoints (BRISK) and Good Features to Track are utilized to identify keypoints in the image. In the second stage, the k-means clustering algorithm is applied to these identified keypoints. The keypoints derived from the good features to track algorithm serve as cluster centers while the keypoints from the BRISK algorithm are distributed around these centers. In the last stage, the logistic map algorithm is employed to add more randomness to the obtained keypoints. This is done by distributing the points using the random list property. The results obtained indicate that the proposed method surpasses comparable techniques in terms of PSNR (Peak Signal-to-Noise Ratio), SSIM (Structural Similarity Index), and BER (Bit Error Rate) values metrics. Thus, the proposed scheme offers a performance advantage over existing methodologies.

Globally, lung cancer remains the leading cause of cancer-related deaths, with early detection significantly improving survival rates. Developing robust machine learning models for early detection necessitates access to high-quality, localized datasets. This project establishes the first lung cancer dataset in Tunisia, utilizing DICOM CT scans from 123 Tunisian patients. The dataset, annotated by experienced radiologists, includes diverse forms of lung cancer at various stages. Using transfer learning with pre-trained 3D ResNet models from Tencent's MedicalNet, our tests showed the dataset outperformed previous models in specificity and sensitivity. This demonstrates its effectiveness in capturing the unique clinical characteristics of the Tunisian population and its potential to significantly enhance lung cancer diagnosis and detection.

Soil erosion is a serious and complex environmental problem worldwide, especially in the centre west of Tunisia. Whereas the construction of hill reservoirs is part of the soil and water conservation strategy, many of these have a siltation problem. Dhkekira is one of the smallest watersheds in central Tunisia whose most lithological formation consists of materials that are quite susceptible to water erosion. Due to the lack of low-scale lithological data, digital IR aerial photos with 2 m spatial resolution were considered. A semi-automatic classification of aerial photos, based on the image's textural indices is developed. The lithologic map extracted from aerial photos was used as input for ANSWERS-2000 water erosion model. Results obtained indicate first, with the semi-automatic classification of the mean and standard deviation of the thumbnail histograms that image output could help to give an idea about the existence of some surface lithological formation. The model applied to Dhkekira watershed showed that the spatial difference in water erosion was not caused only by land cover and slope, but also by lithological formation. The percentage of each lithological formation in sediment yield at the Dhkekira hill reservoir was estimated to be 69% sediment yield from Pleistocene and 19.7% from Lutetian-Priabonian.

Imagery assessment is an efficient method for detecting craniofacial anomalies. A cephalometric landmark matching approach may help in orthodontic diagnosis, craniofacial growth assessment and treatment planning. Automatic landmark matching and anomalies detection helps face the manual labelling limitations and optimize preoperative planning of maxillofacial surgery. The aim of this study was to develop an accurate Cephalometric Landmark Matching method as well as an automatic system for anatomical anomalies classification. First, the Active Appearance Model (AAM) was used for the matching process. This process was achieved by the Ant Colony Optimization (ACO) algorithm enriched with proximity information. Then, the maxillofacial anomalies were classified using the Support Vector Machine (SVM). The experiments were conducted on X-ray cephalograms of 400 patients where the ground truth was produced by two experts. The frameworks achieved a landmark matching error (LE) of 0.50 +/- 1.04 and a successful landmark matching of 89.47% in the 2 mm and 3 mm range and of 100% in the 4 mm range. The classification of anomalies achieved an accuracy of 98.75%. Compared to previous work, the proposed approach is simpler and has a comparable range of acceptable matching cost and anomaly classification. Results have also shown that it outperformed the K-nearest neighbors (KNN) classifier.

This paper presents and evaluates two emerging methods to ensure accurate height measurements under critical conditions. The tests carried out are related to Jack Up Vessel (JUV) foundation height accuracy, and consists in making the appropriate measures in order to achieve a precise installation of an Offshore Wind Farm (OWF), based on Real-Time Kinematic (RTK) and Post-Processed Kinematic (PPK) methods. It is important to note that the OWF project is located far from the coastline, not always in the standard working range of RTK, which makes our task even more complex and restrictive. Again, it is well known in geographic positioning, that the accuracy of vertical coordinates is always more difficult to achieve than the horizontal ones, which explain the complexity of our task to define a precise height measurement of the wind turbines. For the purposes of this study, we use a Global Navigation Satellite System (GNSS) unit referred by DGPS5 (AsteRx-U), a GA830 antenna and Trimble Business Center (TBC) V5.10 software program to make post-processing. To reach a centimeter level of precision, all measures will be evaluated according to the tolerance specification of ±2.5 cm. Based on the test results and analysis, we will show the influence of the baseline lengths to achieve precise measurements and the contribution of the PPK method outside the RTK working range.

In today’s era of escalating digital threats and the growing need for safeguarding sensitive information, this research strives to advance the field of information concealment by introducing a pioneering steganography methodology. Our approach goes beyond the conventional boundaries of image security by seamlessly integrating classical image processing techniques with the cutting-edge realm of quantum encoding. The foundation of our technique lies in the meticulous identification of distinctive features within the cover image, a crucial step achieved through the utilization of SIFT (Scale-Invariant Feature Transform). These identified key points are further organized into coherent clusters employing the K-means clustering algorithm, forming a structured basis for our covert communication process. The core innovation of this research resides in the transformation of the concealed message into a NEQR (Novel Enhanced Quantum Representation) code, a quantum encoding framework that leverages the power of quantum circuits. This transformative step ensures not only the secrecy but also the integrity of the hidden information, making it highly resistant to even the most sophisticated decryption attempts. The strategic placement of the quantum circuit representing the concealed message at the centroids of the clusters generated by the K-means algorithm conceals it within the cover image seamlessly. This fusion of classical image processing and quantum encoding results in an unprecedented level of security for the embedded information, rendering it virtually impervious to unauthorized access. Empirical findings from extensive experimentation affirm the robustness and efficacy of our proposed strategy.

Lung cancer is the most common cancer-related cause of death worldwide. Unfortunately, current diagnostic techniques often lack sensitivity and precision, leading to delayed diagnoses and ineffective treatments. To diagnose lung cancer, doctors currently mainly rely on the clinical characteristics of their patients and imaging characteristics. However, these techniques have limitations in fully and promptly detecting lesions. Nevertheless, with the help of artificial intelligence (AI), lung cancer treatment, prognosis prediction, and diagnostics can be greatly improved. This paper provides an overview of the role that AI can play in simplifying tasks, while reducing the effort required of radiologists and increasing the accuracy of nodule detection.

Autism Spectrum Disorders (ASD) are among the most critical health concerns of our time.These disorders typically present challenges in social interaction, communication, and exhibit repetitive behaviors.To diagnose and customize medical treatments for ASD effectively, the development of robust neuroimaging biomarkers is indispensable.Although extensive studies have recently delved into this area, only a handful have explored the differences between ASD and NC.This study aspires to shed light on this relationship by analyzing both structural and functional brain data associated with ASD.We aim to provide an extensive characterization of ASD by combining techniques of structural and functional analysis.The framework we propose is based on analyzing the differences in structural and functional aspects between ASD and development control (DC) subjects.The study leverages a substantial dataset of 1114 T1-weighted structural and functional Magnetic Resonance Imaging comprising 521 individuals with ASD and 593 controls, ranging in age from 5 to 64 years.These subjects are divided into three broad age categories.Utilizing automated labeling, we compute the features from subcortical and cortical regions.Statistical analyses help identify disparities between ASD and DC subjects.Principal Component Analysis (PCA) is employed to select the most discriminative features, which are subsequently used for classifying the two groups via an Artificial Neural Network (ANN) analysis.Our preliminary findings reveal a significant difference in the distribution of all tested features and subcortical regions between ASD subjects and DC subjects.Through our work, we contribute towards an enhanced understanding of ASD, potentially paving the way for future research and therapeutic interventions.

The aim of this paper is to study and analyze the Jack Up Vessel (JUV) foundation height accuracy, with the objective of the precise installation of an Offshore Wind Farm (OWF), based on Real Time Kinematic (RTK) and Post-Processing Kinematic (PPK) modes applied on short and long baselines length. The offshore wind farm project is located far from the coastline, not always in the standard working range of RTK. The standard allowed vertical installation tolerance for foundations is less than 10 cm. Taking into account all error sources, deformation of the vessel, motion, lever arms that impact the height measurement of the foundation, it is required that RTK and PPK perform within an accuracy less than 5 cm. In this work, all measures will be evaluated according to the tolerance specification of +/- 2.5 cm. The survey GNSS tests executed during the project on board of a JUV should be able to provide answers to the following questions: Despite the critical environment, does RTK method allow reaching the theoretical specifications? Does PPK improve accuracy compared to the RTK solution? What is the influence of the baseline length? How much of the time the results fall within the range tolerance? What is the ideal logging period in which accurate and reliable results can be obtained? What is the influence of the hardware and software variants used in testing process on the results accuracy? Based on the test results and analysis a clear description of the influence of different parameters in the OWF precise height measurement in challenging environment will be exposed.

Erosion is the most dangerous phenomenon of environmental and economic threat to arable land. By FAO. 35 % of Tunisian land are threatened by water erosion and varies regionally. Quantification and estimation of soil loss by water erosion is now essential to install the best management practices. In our study, we used a continuous physical based simulation version of a hydrological and water erosion model ANSWERS-2000. The aim of this study is to investigate the introduction of land use parameters extracted by MODIS image. This study is focused on El Azire watershed, is characterized by a moderate Mediterranean climate and a high spatial heterogeneity of soil and land use proprieties. Describing, monitoring, and predicting land-use and land-cover change in the watershed is a difficult process. However, the usefulness of MODIS image with their daily temporal resolution and spatial resolutions of 250 m and 500 m can make utilization of continuous physical-based models easier for great scale watersheds. A sensitivity analysis was conducted on each application such that the variability in erosion map output and can be assessed and incorporated into the interpretation of results with an acceptable level of confidence.

A new feature extraction approach is proposed in this paper to improve the classification performance in remotely sensed data. The proposed method is based on a primary sources subset (PSS) obtained by nonlinear transform that provides lower space for land pattern recognition. First, the underlying sources are approximated using multilayer neural networks. Given that, Bayesian inferences update unknown sources’ knowledge and model parameters with information’s data. Then, a source dimension minimizing technique is adopted to provide more efficient land cover description. The support vector machine (SVM) scheme is developed by using feature extraction. The experimental results on real multispectral imagery demonstrates that the proposed approach ensures efficient feature extraction by using several descriptors for texture identification and multiscale analysis. In a pixel based approach, the reduced PSS space improved the overall classification accuracy by 13% and reaches 82%. Using texture and multi resolution descriptors, the overall accuracy is 75.87% for the original observations, while using the reduced source space the overall accuracy reaches 81.67% when using jointly wavelet and Gabor transform and 86.67% when using Gabor transform. Thus, the source space enhanced the feature extraction process and allow more land use discrimination than the multispectral observations.

The analysis of the radar response on natural surfaces has been subject of intense research during the last decades in the field of remote sensing. Unless the availability of accurate values of surface roughness parameter, the restitution of soil moisture from radar backscattering signal can constantly provide inaccurate estimates. Characterization of soil roughness is not fully understood, so a wide range of roughness values can be obtained for the same studied surface when using different measurement methodologies. Various studies have shown a weak agreement between experimental measurements of soil physical parameters and theoretical values under natural conditions. Due to this nonlinearity and its ill-posedness, the inversion of backscattering radar signal on soils for restitution of physical soil parameters is particularly complex. The aim of the present work is the restitution of soil physical parameters from backscattered radar signal using an adapted backscattering model to the soil proposed description. As our study focuses on little rough soils, we have adopted in this work a multi-layered modified multiscale bi-dimensional Small Perturbation Model (2D MLS SPM). Subsequently, we propose a new way of describing the dielectric constant, with the aim of including air fractions in the multiscale multilayer description of the soil. Calculating the dielectric constant is based on the consideration of a soil comprising two phases, a fraction of soil, and an air fraction. For the inversion method, a methodology of coupling between neural networks (NN) and genetic algorithms (GA) was carried on in order to restitute the physical properties of the soil. Samples were generated by the original MLS 2D SPM followed by a neural network to obtain the statistic soil moisture and MLS roughness parameters algorithm. Thereafter, these restored values were modelled by the genetic algorithms to resolve, in part or in whole, the disagreement between the retrieval and original values.