Statistics, Optimization & Information Computing

Preface for the Special Issue of International Congress on Engineering and Complex Systems ICECS’23

Mohamed BENSLIMANE — 2024-04-17

The main contributions of [Stat. Optim. Inf. Comput. Vol. , No. (2023)], consisting of seventeen papers selected and revised from the International Congress on Engineering and Complex Systems ICECS’2023, are highlighted.

A Multiobjective Diet Planning Model for Diabetic Patients in the Moroccan Health Context Using Particle Swarm Intelligence

Abdellah Ahourag — 2024-02-21

Recommended diets have a central role to play in creating a healthy dietary environment that enables populations to adopt and maintain health-promoting dietary practices. It’s well known that foods with a low glycemic load (GL) help release a concentration of glucose in the blood, These can contribute to the prevention of various glycemia-elated health problems. We aim to address an optimization model for diets in the Moroccan context that controls both glycemic load and total meal costs. The application of a multiple objective particle swarm method which aggregates the problem’s restrictions with optimized objectives functions helps maintain dietary diversity and facilitates the search for trade-offs between objectives and problem-specific requirements.

A hybrid sampling combining Smote and RF algorithm for cancer chemotherapy protocols Classification

Houda AIT BRAHIM — 2024-02-21

Breast Cancer (BC) is a network of cells that grow abnormally in the breast. If BC is not properly treated with the appropriate cancer chemotherapy protocols, it is at risk of causing death. This research aimed to combine Synthetic Minority Over-sampling (Smote) and Random Forest (RF) methods for BC chemotherapy protocols classification. Smote was used to balance the data, while RF was used to classify chemotherapy protocols data. The real data was produced by collecting medical and personal data from 601 patients with BC at the University Hospital Center (UHC) Mohammed VI of Marrakech in Morocco. Data of women diagnosed with BC from January 2018 at UHC were assessed. The results showed that the use of Smote for data augmentation can increase the performance of the RF classification method based on accuracy. There was an increase of 26% in accuracy. Time is an hyperparameter to be improved.

Multi-Dataset Convolutional Neural Network Model for Glaucoma Prediction in OCT Fundus Scans

Badr ELKARI — 2024-02-21

Glaucoma, a major factor in permanent blindness across the globe, necessitates accurate and efficient diagnostic methods. This article presents a comprehensive approach to glaucoma prediction by combining three diverse datasets: ORIGA, ACRIMA, and REFUGE. A novel multi-Datasets CNN (MD-CNN) architecture is proposed, specifically tailored to effectively handle combined data comprising diverse image characteristics across multiple datasets. This innovative approach demonstrates remarkable robustness in accommodating variations in image attributes, including lighting, zoom levels, and other disparate features, thus showcasing its potency in addressing glaucoma prediction across different datasets. The approach demonstrates improved accuracy (96.88%), sensitivity (94.34%), specificity (97.20%), precision (94.34%), and AUC (99.02%) compared to individual dataset-based models, addressing challenges in glaucoma detection. This research showcases the potential of combining diverse datasets for more effective CNN-based glaucoma detection.

GAN based approaches for self-supervised segmentation: A comparative study

Zohair Elmourabit — 2024-02-21

Image segmentation is a fundamental image processing technique that involves di-viding an image into distinct regions or segments to enable the analysis and extrac-tion of valuable information. It finds applications in various fields including medi-cine, pattern recognition, computer vision, and security surveillance. Different types of image segmentation techniques can be employed based on specific application requirements, including thresholding, region-based segmentation, edge-based segmen-tation, clustering-based segmentation, active contour models (Snakes), graph-based segmentation, watershed segmentation, and deep learning-based segmentation. the latter has become a powerful tool in image segmentation. Deep learning (DL) models can be trained on large datasets of images, allowing them to learn intricate relation-ships between pixels and object classes. This is particularly beneficial for challenging segmentation tasks. However, one significant challenge is the scarcity of labeled training data. To address this issue, self-supervised approaches using generative mod-els like Generative Adversarial Networks (GANs) provide a solution. GANs can gener-ate synthetic training data, which is useful for tasks where acquiring labeled training data is difficult or expensive, such as in medical imaging or remote sensing applica-tions. Additionally, GANs can generate realistic segmentation masks, which are cru-cial for tasks like medical imaging. In this study, we conducted a comparative analy-sis of DL-based segmentation approaches for COVID-19 CT scans. The evaluated approaches were assessed using metrics such as Dice Score, Specificity, and Sensitivi-ty. These metrics provide quantitative measures of the segmentation performance, allowing for an objective evaluation and comparison of the different techniques.

Celestial Object Detection in Astronomical Images Using MSE and Jacobi Moments

Ismail Naouadir — 2024-02-21

Astronomical images share a common characteristic, which is low variability. This term refers to images with minimal changes or variations in pixel values or features. In this article, we leverage the low variability characteristic of astronomical images to detect celestial objects. We compute the mean squared error for the reconstructed astronomical image, derived from its Jacobi moments. These moments are calculated multiple times, focusing on different regions within the image. The mean squared error metric serves as the score function for the simulated annealing algorithm, aiding in the identification of regions with the highest information loss. The parameters used to compute Jacobi moments to focus on that region are then interpreted as the coordinates of celestial objects. This method proves effective for preprocessing images since it provides optimal parameters for Jacobi polynomials, which will enhance their feature extraction capability. Additionally, it serves as an object detection method, as we can interpret the Jacobi moments' parameters as coordinates for the objects within the images.

Optimal image 2D/3D by Krawtchouk moments and ABC algorithm

AAKAM YOUSSEF — 2024-02-21

In our research, we provide an enhanced and effective Krawtchouk moments parameter p optimization approach using the Artificial Bee Colony (ABC) algorithm optimization technique for the purpose of completing tasks including the reconstruction and classification of 3D and 2D images with high quality. The proposed method, which uses the ABC algorithm to generate a suitable parameter based on Krawtchouck moments, aims to accurately characterize the image as a vector known as the descriptor vector of moments, which is then used to reconstruct the image for each order. The simulation and analysis of the findings demonstrate the high level of quality and effectiveness of the suggested KM-ABC approach for image reconstruction, as well as its excellent accuracy in images with and without noise.

License plate text recognition using deep learning, NLP, and image processing techniques

Hanae Moussaoui — 2024-02-21

Detecting license plates has never been easy, particularly with the proliferation of sophisticated radars on highways and roads. By 2021, the gendarmerie and National Security Road control agents will have access to more than 1 billion smart traffic radars worldwide. This research presents a revolutionary technique for detecting and recognizing Arabic and Latin license plates. After assembling the gathered images to create a novel dataset, we utilized YOLO v7 to locate and identify the number plate in the image as the first step of the suggested procedure. Before the dataset was fed to the detection system, it was manually labeled. Afterward, we improved the recognized license plate using machine learning methods. To do this, we used kernel methods as well as thresholding to get rid of the extra vertical lines on the plate. After that, we employed Arabic OCR along with Easy OCR methods to decipher the Latin and Arabic characters on the number plate. Eventually, the proposed method achieved an F1 score of 98%,with a precision and recall of 97% and 98%, respectively. We also obtained an accuracy of 99% for image segmentation. The segmentation and detection results from the suggested strategy have shown satisfactory results.

Securing Color Images with an Innovative Hybrid Method Combining DNA Computing and Chaotic Systems

Hanaa Mansouri — 2024-02-21

Modern cryptography is a key element of data security, ensuring the confidentiality and integrity of information. In an increasingly digital world, cryptography remains crucial for the protection of sensitive data. In this context, we propose a novel hybrid security system for encrypted color images using a DNA model, chaotic systems, and SHA256-MD5 hash functions as a basis. The proposed hybrid system includes DNA permutation and diffusion. In DNA permutation, we unpredictably rearrange the location of DNA image elements by using the logistic map of low computational complexity. In DNA diffusion, we diffuse the permuted image of DNA with the DNA image key generated by a 5D hyper-chaotic system, using a variety of algebraic operators such as the circular offset in both directions. Considering the experimental outcomes and security evaluation, we can infer that the proposed hybrid security system demonstrates a high level of security, resistance to existing attacks, and practical application suitability while maintaining speed.

Finite Element modeling and convergence analysis of a new Biomimetic Branching Structure.

Nadir RIHANI — 2024-02-21

Branching structures are gaining popularity in the field of advanced structures and building design; they offer high performance in terms of strength and lightweight design, along with the flexibility and precision enabled by modern processing technologies like Additive Manufacturing. This paper provides a concise overview of a geometric design procedure for a novel ribbed class of structures which was previously developed by the authors as a biomimetic optimal Micro-architected dome. Hereinafter, linear lattice models are suggested to carry out structural calculations using the finite element method (FEM). The objective is to examine discretization thresholding and strain energy convergence criteria. Results show that convergence is reached for numbers of elements per leg, ranging from 2 to 6, depending on the geometrical configuration of the dome being studied. The strain energy balance also exposes the influence of each internal force on the total mechanical response of the structure, pinpointing bending moment and axial force as the main decisive factors. As a perspective, the study will focus on limit state design calculations and Analysis of how the local geometry influences the overall stability and strength of this new design.

Numerical simulation study of the effect of integrating hemp concrete and passive strategies on the energy consumption of a residential building in Al-Hoceima

Hicham Kaddouri — 2024-02-21

In this work, we present an energy efficiency study for a residential building located in the city of Al-Hoceima, Morocco. The aim is to bring the building into compliance with the technical requirements of the Moroccan Thermal Building Regulations (RTCM). To achieve this, several modifications and interventions were carried out, such as the use of hemp concrete for insulation and the integration of passive strategies to minimize energy loads. Similarly, this energy study is carried out as a simulation using TRNSYS software, so that the building is modeled as a multizone entity with an occupancy schedule. The simulation results show that the technical requirements of the RTCM are achieved using a construction containing hemp concrete with an optimum thickness of around 10cm, this thickness can be reduced to 4cm using double glazing, or to 1cm using controlled natural ventilation in summer and winter.

Assessing public street lighting in Morocco: evaluating efficiency and the impact of geometrical parameters on luminance

Younes AGROUAZ — 2024-02-21

The main objective of this study is to present a bench marking of the public street lighting in typical road in Morocco. The study deals with a comparison between the conventional technology used in Moroccan road and the LED technology. The simulated system is based on road with two directions and a sidewalk, three arrangements type of pole layout are treated (Left, Middle and Two sides arrangement). Also, in this paper deals with a parametric analysis based on the variation of boom angle, light Center high and the pole distance in order to determine the luminance of each arrangement at each pole distance. The simulation is carried out using two luminaries, the fi rst one represents a conventional technology and LED lamp. The founding of this paper shows that the LED lamp present a high luminance value of 20 cd/m² for a pole distance of 10m and boom angle of 0°, in comparison with the conventional lamp that present a luminance of 8 cd/m² at the same condition. The left arrangement type shows the best energy effi ciency indicators such as DE and power per km, this means that this disposition is less consuming of energy and acceptable luminance values. Therefore, these two indicators play a crucial role in enhancing the effi ciency of the lighting.

Modeling and optimization of controlled landfill gas performance using LandGEM

OUISSAL DRISSI EL BOUZAIDI — 2024-02-21

Modeling with LandGEM offers a large field to quantify emissions from the decomposition of organic waste, consequently prediction of energy generated from gas emissions especially those of methane. Where, the simulation results makes it possible to define the reasons of malfunction concluded by comparing between the LandGEM simulation and the experimental methodology for estimation of methane gas emissions, whose offer a huge knowing, accuracy assessment, and detailed analysis to landfill site state in terms of the production of methane, the tightness, the progress of the anaerobic degradation, the existence of sulfur compound. This study accomplished in site the methane production at 80%, a tightness of 84% compared to the 100% given by LandGEM, and an hydrogen sulfide content which far exceed the recommended limit, cause of corrosion attacking especially the biogas compressor which provoke a complete shutdown to product the electrical energy. Unlike the LandGEM digital model, whose site must produce 3,145.107 m3 of methane with 100% site tightness, a hydrogen sulfide content equivalent to 40 ppm, for a production of electrical energy equivalent to 10,941.107kWh according to the year ‘2022’. To optimizing landfill gas performance, we forecast the installation of automatic analyzer sensors to anticipate of an alert for intervention, generated by accidental loss of methane in the air, infiltration of oxygen inside the lockers, disruption of the stability of the fermentation process, or existence of high sulfur contents in organic matter.

Exploring Maze Navigation: A Comparative Study of DFS, BFS, and A* Search Algorithms.

Badr ELKARI — 2024-02-21

A comparative study was conducted using Python and Pmaze to evaluate the performance of DFS, BFS, and A* search algorithms in path planning for a maze. The main objective of the project was to compare the efficiency of these algorithms in terms of path cost and algorithmic complexity. No physical robot was used in this study as it focused solely on the analysis of search algorithms. This study contributes to the advancement of knowledge in the field of path planning by exploring different approaches for mazes using Python and Pmaze.

Object Knowledge Integration for Skeleton-Based Action Recognition

Oumaima Moutik — 2024-02-21

3D-Skeleton-based action recognition has been widely adopted due to its efficiency and robustness to complex backgrounds. While it is capable of conveying a significant amount of information regarding the dynamics of human poses, we argue that its performance is curtailed when confronted with actions involving interactions between humans and objects due to the absence of the study of the surrounding objects. It is of great importance to delve deeper into the study of human-object interactions for skeleton-based action recognition. This paper proposes a novel approach to represent the spatial-temporal skeleton features, along with the present nearby objects and their dynamics. To accomplish this, a new formulation named object knowledge is presented, which entails the categorization of object characteristics, based on whether or not the object necessitates a motion analysis. With a piece of prior knowledge, in cases where it is required, the motion is calculated, while for those where it is not necessary, only the category of object is considered. This object knowledge is then early-fusion along with the skeleton representation, in such a way that it fits into the self-attention model. The experimental results on different popular action recognition datasets (NTU RGB+D 60, NTU RGB+ D 120 illustrate that the proposed approach outperforms the current state-of-the-art methods.

Portfolio selection problem: main knowledge and models (A systematic review)

Mohammed Ziane — 2024-02-21

The challenge in portfolio optimization lies in creating a collection of assets that attains a target expected return
while mitigating risk. This problem is often framed as an optimization task, specifically Mean Variance Optimization (MVO). MVO involves formulating an objective function, typically quadratic, that is contingent on the composition of the portfolio, and linear constraints that represents the portfolio's asset allocation restriction. Several improvements have been proposed, such as adding constraints to MVO or using alternative risk measures. As a result, even though MVO model remains the most widely studied type of portfolio optimization, different types of portfolio optimization models, risk/return measurements and constraints have been suggested and used since its invention. In this work, we delve into the various risk and return measures, constraints, and mathematical models commonly used in portfolio optimization.We discuss the key risk measures employed in portfolio optimization, including the Sharpe ratio, beta, maximum drawdown and others, We explore the constraints commonly applied in portfolio optimization. Furthermore, we delve into the mathematical models utilized in portfolio optimization. Then, we emphasize the interplay between risk and return measures, constraints, and mathematical models in portfolio optimization. By providing a comprehensive overview of risk and return measures, constraints, and mathematical models, this work aims to enhance the understanding of portfolio optimization techniques and facilitate informed decisionmaking in the field of investment management. To illustrate different knowledge and models, several experiments were conducted on well-known real data portfolios.

Economic growing quantity: Broiler chicken in Morocco

AZEDINE OUHMID — 2024-02-21

In this work, we introduce an innovative mathematical model for growing items, with a particular emphasis on broiler chicken production within the context of Moroccan agriculture. The main goal of this study is to determine the optimal order quantity of items that the inventory should purchase to satisfy customer demand and the corresponding cycle duration. For the first time, a staircase function is used to depict the gradual reduction in item quantities over time due to the mortality during the rearing period. Furthermore, and for the first time, our model incorporates three distinct feeding types—Starter Feed, Grower Feed, and Finisher Feed-bringing it in line with real-world agricultural practices. A numerical example is provided, based on data extended from a Moroccan farm project, along with an analytic solution.

Optimal reconstruction and recognition of images by Jacobi Fourier moments and artificial bee colony (ABC) algorithm

SAHMOUDI Yahya — 2024-02-21

The orthogonal moments giving relevant results of these last years within the framework of object detection, pattern recognition and image reconstruction, this work based on orthogonal functions called Orthogonal Jacobi Polynomials (OJPs), and we introduce a new set of moments called Generalized Jacobi Fourier Moments (GJFMs), these polynomials are characterized by parameters . However, it was very important to optimize these parameters in order to obtain a good result, in this context; this study used a new approach to optimized Jacobi Fourier parameters using the artificial bee colony algorithm (ABC) in order to improves the quality of reconstruction of images of large sizes. On the one hand, to validate this technique which offers a high image reconstruction quality. On other hand, the comparison carried out with other algorithms clearly indicates the advantage of the proposed method.