The advancement of deep learning methods across various applications has forced the creation of enormous training datasets. However, obtaining suitable real-world datasets is often challenging for various reasons. Consequently, numerous studies have emerged focusing on the generation and utilization of synthetic data in the training process. Hence, there is no universal formula for preparing synthetic data and leveraging it in network training to maximize the effectiveness of various detection methods. This work provides a comprehensive overview of several synthetic data generation techniques, followed by a thorough investigation into the impact of training methods and the selection of synthetic data quantities. The outcomes of this research enable the formulation of conclusions regarding the recipe for developing synthetic data with high efficacy in enhancing detection methods. The main conclusion for the synthetic data generation methods is to ensure maximum diversity at a high level of photorealism, which allows improving the classification quality by more than 5% to even 19% for different detection metrics.

The application of image analysis methods to calculate the distance from the camera to the object allows the replacement of specialized hardware devices for distance estimation. In the case of public transport, estimation of the exact position of the passenger gives also the option to determine whether the passenger is inside or outside the vehicle. In the presented work, several distance estimation methods based on typical analytical models and machine learning (ML) methods were tested using recordings from three cameras located in the minibus model. Human head detection was used instead of the entire passenger body to avoid occlusion problems. The analytical method showed worse performance than ML methods in all cases. The difference in the performance of ML models between cameras was negligible and there was no best method found. The computational time for ML models ranges from 0.35 to 100.57 ms, which should result in successful real-world applications. The developed approach can be used not only in public transport but also in all closed areas for the calculation of people or crowd density.

Videodermoscopy is the gold standard in melanoma diagnosis. A new research trend aims to support specialists' work by automating this examination method, but most research focuses on developing detectors to replace dermatologists. Our study aims to present a novel idea which focuses on supporting the examination process instead of completely automating it. Instead of detecting and classifying all naevi each time, our approach focuses on full screening at the first visit and shortening the examination at subsequent visits. During the following examinations, images of all naevi are taken by supporting staff and are compared with previously taken images using feature-matching or co-registration-based algorithms. An unmatched naevus needs to be checked by the specialist, meaning that a new naevus appeared or an old one had changed. The similarity between images (ranging from 0% to 100%) was obtained with characteristic points searching (ORB method) and co-registration-based methods. No dataset presenting changes of naevi over time exists, so augmenting the images from PH2 and HAM10000 datasets was required to simulate the time flow. The decision bound was set, so the true positive rate of each method was equal to 99%, giving the specificity of ORB equal to 97% (PH2) and 89% (HAM10000). Similar tests were performed using a co-registration algorithm giving specificity equal to 23% (PH2) and 19% (HAM10000). We believe that employing our approach will allow supporting staff to perform screening without experts' help, decreasing examination time and costs and increasing the accessibility of dermoscopic examinations.