License Plate Images Generation with Diffusion Models

Despite the evident practical importance of license plate recognition (LPR), corresponding research is limited by the volume of publicly available datasets due to privacy regulations such as the General Data Protection Regulation (GDPR). To address this challenge, synthetic data generation has emerged as a promising approach. In this paper, we propose to synthesize realistic license plates (LPs) using diffusion models, inspired by recent advances in image and video generation. In our experiments a diffusion model was successfully trained on a Ukrainian LP dataset, and 1000 synthetic images were generated for detailed analysis. Through manual classification and annotation of the generated images, we performed a thorough study of the model output, such as success rate, character distributions, and type of failures. Our contributions include experimental validation of the efficacy of diffusion models for LP synthesis, along with insights into the characteristics of the generated data. Furthermore, we have prepared a synthetic dataset consisting of 10,000 LP images, publicly available at https://zenodo.org/doi/10.5281/zenodo.13342102. Conducted experiments empirically confirm the usefulness of synthetic data for the LPR task. Despite the initial performance gap between the model trained with real and synthetic data, the expansion of the training data set with pseudolabeled synthetic data leads to an improvement in LPR accuracy by 3% compared to baseline.

Electrostatic Force Regularization for Neural Structured Pruning

The demand for deploying deep convolutional neural networks (DCNNs) on resource-constrained devices for real-time applications remains substantial. However, existing state-of-the-art structured pruning methods often involve intricate implementations, require modifications to the original network architectures, and necessitate an extensive fine-tuning phase. To overcome these challenges, we propose a novel method that, for the first time, incorporates the concepts of charge and electrostatic force from physics into the training process of DCNNs. The magnitude of this force is directly proportional to the product of the charges of the convolution filter and the source filter, and inversely proportional to the square of the distance between them. We applied this electrostatic-like force to the convolution filters, either attracting filters with opposite charges toward non-zero weights or repelling filters with like charges toward zero weights. Consequently, filters subject to repulsive forces have their weights reduced to zero, enabling their removal, while the attractive forces preserve filters with significant weights that retain information. Unlike conventional methods, our approach is straightforward to implement, does not require any architectural modifications, and simultaneously optimizes weights and ranks filter importance, all without the need for extensive fine-tuning. We validated the efficacy of our method on modern DCNN architectures using the MNIST, CIFAR, and ImageNet datasets, achieving competitive performance compared to existing structured pruning approaches.

2D and 3D CNN-Based Fusion Approach for COVID-19 Severity Prediction from 3D CT-Scans

Since the appearance of Covid-19 in late 2019, Covid-19 has become an active research topic for the artificial intelligence (AI) community. One of the most interesting AI topics is Covid-19 analysis of medical imaging. CT-scan imaging is the most informative tool about this disease. This work is part of the 3nd COV19D competition for Covid-19 Severity Prediction. In order to deal with the big gap between the validation and test results that were shown in the previous version of this competition, we proposed to combine the prediction of 2D and 3D CNN predictions. For the 2D CNN approach, we propose 2B-InceptResnet architecture which consists of two paths for segmented lungs and infection of all slices of the input CT-scan, respectively. Each path consists of ConvLayer and Inception-ResNet pretrained model on ImageNet. For the 3D CNN approach, we propose hybrid-DeCoVNet architecture which consists of four blocks: Stem, four 3D-ResNet layers, Classification Head and Decision layer. Our proposed approaches outperformed the baseline approach in the validation data of the 3nd COV19D competition for Covid-19 Severity Prediction by 36%.

Dynamic camera alignment optimization problem based on Fractal Decomposition based Algorithm

In this work, we tackle the Dynamic Optimization Problem (DOP) of IA in a real-world application using a Dynamic Optimization Algorithm (DOA) called Fractal Decomposition Algorithm (FDA), introduced by recently. We used FDA to perform IA on CCTV camera feed from a tunnel. As the camera viewpoint can change by multiple reasons such as wind, maintenance, etc. the alignment is required to guarantee the correct functioning of video-based traffic security system.

Graph based adaptive evolutionary algorithm for continuous optimization

he greatest weakness of evolutionary algorithms, widely used today, is the premature convergence due to the loss of population diversity over generations. To overcome this problem, several algorithms have been proposed, such as the Graph-based Evolutionary Algorithm (GEA) \cite{1} which uses graphs to model the structure of the population, but also memetic or differential evolution algorithms \cite{2,3}, or diversity-based ones \cite{4,5} have been designed. These algorithms are based on multi-populations, or often rather focus on the self-tuning parameters, however, they become complex to tune because of their high number of parameters. In this paper, our approach consists of an evolutionary algorithm that allows a dynamic adaptation of the search operators based on a graph in order to limit the loss of diversity and reduce the design complexity.

Multi-node environment strategy for Parallel Deterministic Multi-Objective Fractal Decomposition

This paper presents a new implementation of deterministic multiobjective (MO) optimization called Multiobjective Fractal Decomposition Algorithm (Mo-FDA). The original algorithm was designed for mono-objective large scale continuous optimization problems. It is based on a divide and conquer strategy and a geometric fractal decomposition of the search space using hyperspheres. Then, to deal with MO problems a scalarization approach is used. In this work, a new approach has been developed on a multi-node environment using containers. The performance of Mo-FDA was compared to state of the art algorithms from the literature on classical benchmark of multi-objective optimization