Louis Aimé Fono
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Louis Aimé Fono has a total of 48 co-authors. Here are the top co-authors by number of shared publications:
- Jules Sadefo Kamdem
MRE UR 209 and Faculty of Economics, Montpellier University, Montpellier, France
8 shared publications - Gaston Clément
Laboratoire de Mathématique, Université de Douala, Douala, Cameroon
4 shared publications - Auguste Vigny Noumsi Woguia
University of Douala
2 shared publications - Christian Deffo Tassak
Laboratoire de Mathématiques Appliquées aux Sciences Sociales Université de Yaoundé I B.P. 812 Yaoundé Cameroon
2 shared publications - Joseph Mvogo Ngono
Laboratory of Applied Computer Science, Department of Computer Science, Faculty of Science, University of Douala, Douala, Cameroon
2 shared publications Louis Fono
Laboratoire de Mathématiques et Faculté des Sciences Université de Douala B.P. 24157 Douala
2 shared publications
Latest Publications
- Open Access
- 19 January 2026
- Published by MDPI AG in Applied Sciences
- Vol. 16 (2) , 986
Eye Gaze Detection Using a Hybrid Multimodal Deep Learning Model for Assistive Technology
This paper presents a novel hybrid multimodal deep learning model for robust and real-time eye gaze estimation. Accurate gaze tracking is essential for advancing human–computer interaction (HCI) and assistive technologies, but existing methods often struggle with environmental variations, require extensive calibration, and are computationally intensive. Our proposed model, GazeNet-HM, addresses these limitations by synergistically fusing features from RGB, depth, and infrared (IR) imaging modalities. This multimodal approach allows the model to leverage complementary information: RGB provides rich texture, depth offers invariance to lighting and aids pose estimation, and IR ensures robust pupil detection. Furthermore, we introduce a personalized adaptation module that dynamically fine-tunes the model to individual users with minimal calibration data. To ensure practical deployment, we employ advanced model compression techniques, enabling real-time inference on resource-constrained embedded systems. Extensive evaluations on public datasets (MPIIGaze, EYEDIAP, Gaze360) and our collected M-Gaze dataset demonstrate that GazeNet-HM achieves state-of-the-art performance, reducing the mean angular error by up to 27.1% compared to leading unimodal methods. After model compression, the system achieves a real-time inference speed of 32 FPS on an embedded Jetson Xavier NX platform. Ablation studies confirm the contribution of each modality and component, highlighting the effectiveness of our holistic design.
- 19 January 2026
- Published by MDPI AG in Applied Sciences
- Vol. 16 (2) , 986
Open AccessEye Gaze Detection Using a Hybrid Multimodal Deep Learning Model for Assistive Technology
This paper presents a novel hybrid multimodal deep learning model for robust and real-time eye gaze estimation. Accurate gaze tracking is essential for advancing human–computer interaction (HCI) and assistive technologies, but existing methods often struggle with environmental variations, require extensive calibration, and are computationally intensive. Our proposed model, GazeNet-HM, addresses these limitations by synergistically fusing features from RGB, depth, and infrared (IR) imaging modalities. This multimodal approach allows the model to leverage complementary information: RGB provides rich texture, depth offers invariance to lighting and aids pose estimation, and IR ensures robust pupil detection. Furthermore, we introduce a personalized adaptation module that dynamically fine-tunes the model to individual users with minimal calibration data. To ensure practical deployment, we employ advanced model compression techniques, enabling real-time inference on resource-constrained embedded systems. Extensive evaluations on public datasets (MPIIGaze, EYEDIAP, Gaze360) and our collected M-Gaze dataset demonstrate that GazeNet-HM achieves state-of-the-art performance, reducing the mean angular error by up to 27.1% compared to leading unimodal methods. After model compression, the system achieves a real-time inference speed of 32 FPS on an embedded Jetson Xavier NX platform. Ablation studies confirm the contribution of each modality and component, highlighting the effectiveness of our holistic design.
- Open Access
- 1 January 2026
- Published by Wiley in Advances in Human-Computer Interaction
- Vol. 2026 (1) , 3300666
TinyGaze v2: A Systems Benchmark for Practical Gaze Estimation on Edge Hardware
The deployment of robust, real‐time gaze estimation on commodity hardware is critical for pervasive gaze‐aware augmented and virtual reality (AR/VR). While the computer vision community has made significant progress in accuracy, a pervasive focus on leaderboard performance has created a gap between research and the system‐level constraints of real‐world edge devices. High‐end smartphones and AR/VR headsets increasingly incorporate dedicated ASIC‐based eye tracking hardware, yet such solutions are not universally accessible, as they require custom sensor arrays, fixed camera placements and proprietary SDKs that preclude cross‐platform deployment. This paper presents TinyGaze v2, an open‐source benchmarking suite for edge‐based gaze estimation. We introduce a novel set of hardware‐centric metrics (FPS‐at‐5W, Sustained FPS) and a standardised evaluation protocol. As a concrete instantiation of this benchmark, we provide an optimised gaze estimation model that achieves 58 FPS at 4.8° error on a Jetson Nano while consuming under 4 W. Through rigorous evaluation on a diverse hardware testbed (Jetson Nano, Raspberry Pi + TPU, Google Pixel 7), our multiobjective bootstrap analysis shows that TinyGaze v2 and FullFace are the only statistically nondominated models, occupying complementary Pareto regimes: FullFace for highest accuracy (4.10°, 8 FPS) and TinyGaze v2 for highest efficiency (4.81°, 58 FPS). Our work provides the community with the necessary tools and perspective—not a single model, but a reproducible benchmarking framework—to build feasible, real‐world gaze‐enabled systems.
- 1 January 2026
- Published by Wiley in Advances in Human-Computer Interaction
- Vol. 2026 (1) , 3300666
Open AccessTinyGaze v2: A Systems Benchmark for Practical Gaze Estimation on Edge Hardware
The deployment of robust, real‐time gaze estimation on commodity hardware is critical for pervasive gaze‐aware augmented and virtual reality (AR/VR). While the computer vision community has made significant progress in accuracy, a pervasive focus on leaderboard performance has created a gap between research and the system‐level constraints of real‐world edge devices. High‐end smartphones and AR/VR headsets increasingly incorporate dedicated ASIC‐based eye tracking hardware, yet such solutions are not universally accessible, as they require custom sensor arrays, fixed camera placements and proprietary SDKs that preclude cross‐platform deployment. This paper presents TinyGaze v2, an open‐source benchmarking suite for edge‐based gaze estimation. We introduce a novel set of hardware‐centric metrics (FPS‐at‐5W, Sustained FPS) and a standardised evaluation protocol. As a concrete instantiation of this benchmark, we provide an optimised gaze estimation model that achieves 58 FPS at 4.8° error on a Jetson Nano while consuming under 4 W. Through rigorous evaluation on a diverse hardware testbed (Jetson Nano, Raspberry Pi + TPU, Google Pixel 7), our multiobjective bootstrap analysis shows that TinyGaze v2 and FullFace are the only statistically nondominated models, occupying complementary Pareto regimes: FullFace for highest accuracy (4.10°, 8 FPS) and TinyGaze v2 for highest efficiency (4.81°, 58 FPS). Our work provides the community with the necessary tools and perspective—not a single model, but a reproducible benchmarking framework—to build feasible, real‐world gaze‐enabled systems.
- 18 December 2025
- Published by World Scientific Pub Co Pte Ltd in International Journal of Reliability, Quality and Safety Engineering
- Vol. 33 (02)
On the (Hoang) Pham Distribution
The Pham distribution [H. Pham, A vtub-shaped hazard rate function with applications to system safety, Int. J. Reliab. Appl. 3 (2002) 1–16] has not been studied much in spite of being introduced over 20 years ago. Several of it basic mathematical properties are not known in closed form. In this paper, explicit expressions are derived for its mathematical properties, including hazard rate function, reversed hazard rate function, shapes of the probability density, hazard rate and reversed hazard rate functions, stochastic orderings, moment properties, quantiles, mean deviations, mean residual life, expected inactive time, Lorenz, Bonferroni and Zenga curves, order statistics, extreme value statistics, geometric mean, Shannon entropy, Rényi entropy, Tsallis entropy, cumulative residual entropy, maximum likelihood estimation, observed information matrix and expected information matrix. A simulation study is performed to assess finite sample performance of the maximum likelihood estimators. A data application of the Pham distribution is illustrated.
- 18 December 2025
- Published by World Scientific Pub Co Pte Ltd in International Journal of Reliability, Quality and Safety Engineering
- Vol. 33 (02)
On the (Hoang) Pham Distribution
The Pham distribution [H. Pham, A vtub-shaped hazard rate function with applications to system safety, Int. J. Reliab. Appl. 3 (2002) 1–16] has not been studied much in spite of being introduced over 20 years ago. Several of it basic mathematical properties are not known in closed form. In this paper, explicit expressions are derived for its mathematical properties, including hazard rate function, reversed hazard rate function, shapes of the probability density, hazard rate and reversed hazard rate functions, stochastic orderings, moment properties, quantiles, mean deviations, mean residual life, expected inactive time, Lorenz, Bonferroni and Zenga curves, order statistics, extreme value statistics, geometric mean, Shannon entropy, Rényi entropy, Tsallis entropy, cumulative residual entropy, maximum likelihood estimation, observed information matrix and expected information matrix. A simulation study is performed to assess finite sample performance of the maximum likelihood estimators. A data application of the Pham distribution is illustrated.
- 8 July 2025
- Published by Springer Nature in Computational Economics
- Vol. 67 (1) , 253-277
Mixed Modified Fractional Merton Model of the Bear Spread Basket Put Option using the Multidimensional Mellin Transform
- 8 July 2025
- Published by Springer Nature in Computational Economics
- Vol. 67 (1) , 253-277
Mixed Modified Fractional Merton Model of the Bear Spread Basket Put Option using the Multidimensional Mellin Transform
- 18 June 2025
- Published by World Scientific Pub Co Pte Ltd in International Journal of Financial Engineering
- Vol. 12 (02)
Interest rate options in one-factor mixed modified fractional Vasicek model
This paper introduces an innovative interest rate model in which the dynamics of each factor influencing the underlying short-term interest rate is described using a mixed modified fractional Vasicek framework. The study focuses on (a) formulating the short-term interest rate model and identifying its key parameters, including expectations, variance, and the variance–covariance matrix of the model factors, and (b) computing the integral of this novel model to derive zero-coupon bond prices and corresponding yield expressions across maturities, enabling the construction of a robust term structure model based on mixed modified fractional motion. The model is applied to yield curve data from selected Central African countries, including Cameroon, Congo, and Gabon. This involves (i) presenting the observed yield curve data, (ii) estimating model parameters to align with the theoretical framework, and (iii) reconstructing the latent states (underlying factors) of the interest rate model using the Kalman filter algorithm. The comparative analysis between the latent states simulated by the Kalman filter and the theoretical model highlights the algorithm’s efficiency in parameter optimization. Furthermore, the empirical validation in two stages substantiates the robustness and applicability of the model to capture the interest rate dynamics of Central African economies. Beyond its regional focus, this novel framework demonstrates versatility, offering potential applications in modeling interest rate behavior across other emerging and OECD economies.
- 18 June 2025
- Published by World Scientific Pub Co Pte Ltd in International Journal of Financial Engineering
- Vol. 12 (02)
Interest rate options in one-factor mixed modified fractional Vasicek model
This paper introduces an innovative interest rate model in which the dynamics of each factor influencing the underlying short-term interest rate is described using a mixed modified fractional Vasicek framework. The study focuses on (a) formulating the short-term interest rate model and identifying its key parameters, including expectations, variance, and the variance–covariance matrix of the model factors, and (b) computing the integral of this novel model to derive zero-coupon bond prices and corresponding yield expressions across maturities, enabling the construction of a robust term structure model based on mixed modified fractional motion. The model is applied to yield curve data from selected Central African countries, including Cameroon, Congo, and Gabon. This involves (i) presenting the observed yield curve data, (ii) estimating model parameters to align with the theoretical framework, and (iii) reconstructing the latent states (underlying factors) of the interest rate model using the Kalman filter algorithm. The comparative analysis between the latent states simulated by the Kalman filter and the theoretical model highlights the algorithm’s efficiency in parameter optimization. Furthermore, the empirical validation in two stages substantiates the robustness and applicability of the model to capture the interest rate dynamics of Central African economies. Beyond its regional focus, this novel framework demonstrates versatility, offering potential applications in modeling interest rate behavior across other emerging and OECD economies.
- 13 May 2025
- Published by Springer Nature in Environmental Modeling & Assessment
- Vol. 30 (5) , 947-962
Impact of Human Activities and Forest Biomass on Atmospheric Carbon Dioxide: A Mathematical Model
- 13 May 2025
- Published by Springer Nature in Environmental Modeling & Assessment
- Vol. 30 (5) , 947-962
Impact of Human Activities and Forest Biomass on Atmospheric Carbon Dioxide: A Mathematical Model
- Open Access
Characterization of a coherent fuzzy risk measure
This paper provides new mathematical tools to describe and quantify fuzzy risk in some economic context mainly impacted by fuzzy random phenomena. More precisely, the concept of fuzzy risk measure is defined and characterized in order to deal with randomness and impreciseness in some economic areas. Furthermore, we implement the fuzzy risk measure to solve risk capital allocation problems by considering the α-consensus value introduced and characterized by Nan et al.[20] as the solution concept.
Open AccessCharacterization of a coherent fuzzy risk measure
This paper provides new mathematical tools to describe and quantify fuzzy risk in some economic context mainly impacted by fuzzy random phenomena. More precisely, the concept of fuzzy risk measure is defined and characterized in order to deal with randomness and impreciseness in some economic areas. Furthermore, we implement the fuzzy risk measure to solve risk capital allocation problems by considering the α-consensus value introduced and characterized by Nan et al.[20] as the solution concept.
- Open Access
- 14 February 2025
- Published by Elsevier in Data in Brief
- Vol. 59, 111394
Towards precision agriculture: A dataset for early detection of corn leaf pests
- 14 February 2025
- Published by Elsevier in Data in Brief
- Vol. 59, 111394
Open AccessTowards precision agriculture: A dataset for early detection of corn leaf pests
- Open Access
- 1 January 2024
- Published by IGI Global Scientific Publishing in International Journal of Information Security and Privacy
- Vol. 18 (1) , 1-17
Leveraging TenSEAL
In our interconnected world, the use of Internet of Things (IoT) devices generates vast amounts of sensitive data. To secure this data while extracting insights, privacy-preserving machine learning (PPML) tools like Fully Homomorphic Encryption (FHE) are crucial. This study compares two FHE schemes, BFV and CKKS, for protecting IoT data during PPML analysis. Logistic regression models are implemented on encrypted datasets using the TenSEAL library, and the accuracy and efficiency are compared with unencrypted data. The study focuses on privacy preservation, computational overhead, and model accuracy. Using the capabilities of TenSEAL, the study finds that both BFV and CKKS effectively preserve IoT data privacy during ML tasks. BFV demonstrates superior computational efficiency, enabling faster analysis and training, while CKKS achieves higher accuracy in predicting IoT data. This study aids researchers in selecting the most suitable homomorphic encryption scheme for their needs.
- 1 January 2024
- Published by IGI Global Scientific Publishing in International Journal of Information Security and Privacy
- Vol. 18 (1) , 1-17
Open AccessLeveraging TenSEAL
In our interconnected world, the use of Internet of Things (IoT) devices generates vast amounts of sensitive data. To secure this data while extracting insights, privacy-preserving machine learning (PPML) tools like Fully Homomorphic Encryption (FHE) are crucial. This study compares two FHE schemes, BFV and CKKS, for protecting IoT data during PPML analysis. Logistic regression models are implemented on encrypted datasets using the TenSEAL library, and the accuracy and efficiency are compared with unencrypted data. The study focuses on privacy preservation, computational overhead, and model accuracy. Using the capabilities of TenSEAL, the study finds that both BFV and CKKS effectively preserve IoT data privacy during ML tasks. BFV demonstrates superior computational efficiency, enabling faster analysis and training, while CKKS achieves higher accuracy in predicting IoT data. This study aids researchers in selecting the most suitable homomorphic encryption scheme for their needs.
- Open Access
- 1 January 2024
- Published by Walter de Gruyter GmbH in Journal of Intelligent Systems
- Vol. 33 (1)
A robust segmentation method combined with classification algorithms for field-based diagnosis of maize plant phytosanitary state
Early diagnosis of maize-plant phytosanitary state in the field is crucial to prevent crop damage and optimize yield. However, this field diagnosis presents a challenge due to the variable background of the field environment, which can hinder the performance of classification algorithms. In this article, we introduced a novel segmentation technique using a combined normalized difference vegetation index that effectively isolates the features of interest, such as the leaves, from the surrounding image, which includes the diverse field background. To assess the effectiveness of our segmentation approach, we conducted early diagnosis of maize plants in the field using supervised classification algorithms. We generated a dataset that incorporated four essential texture features: energy, entropy, contrast, and inverse. These features were extracted from each of the segmented images using grayscale co-occurrence matrices. We employed four different classification methods, namely Adaboost, Random Forest, K-Nearest Neighbors, and support vector machine. When combined with the proposed segmentation technique, the support vector machine outperformed the other models, achieving an accuracy rate of 97%.
- 1 January 2024
- Published by Walter de Gruyter GmbH in Journal of Intelligent Systems
- Vol. 33 (1)
Open AccessA robust segmentation method combined with classification algorithms for field-based diagnosis of maize plant phytosanitary state
Early diagnosis of maize-plant phytosanitary state in the field is crucial to prevent crop damage and optimize yield. However, this field diagnosis presents a challenge due to the variable background of the field environment, which can hinder the performance of classification algorithms. In this article, we introduced a novel segmentation technique using a combined normalized difference vegetation index that effectively isolates the features of interest, such as the leaves, from the surrounding image, which includes the diverse field background. To assess the effectiveness of our segmentation approach, we conducted early diagnosis of maize plants in the field using supervised classification algorithms. We generated a dataset that incorporated four essential texture features: energy, entropy, contrast, and inverse. These features were extracted from each of the segmented images using grayscale co-occurrence matrices. We employed four different classification methods, namely Adaboost, Random Forest, K-Nearest Neighbors, and support vector machine. When combined with the proposed segmentation technique, the support vector machine outperformed the other models, achieving an accuracy rate of 97%.