Currently there is no publicly available adequate dataset that could be used for training Generative Adversarial Networks (GANs) on car images. All available car datasets differ in noise, pose, and zoom levels. Thus, the objective of this work was to create an improved car image dataset that would be better suited for GAN training. To improve the performance of the GAN, we coupled the LSUN and Stanford car datasets. A new merged dataset was then pruned in order to adjust zoom levels and reduce the noise of images. This process resulted in fewer images that could be used for training, with increased quality though. This pruned dataset was evaluated by training the StyleGAN with original settings. Pruning the combined LSUN and Stanford datasets resulted in 2,067,710 images of cars with less noise and more adjusted zoom levels. The training of the StyleGAN on the LSUN-Stanford car dataset proved to be superior to the training with just the LSUN dataset by 3.7% using the Fréchet Inception Distance (FID) as a metric. Results pointed out that the proposed LSUN-Stanford car dataset is more consistent and better suited for training GAN neural networks than other currently available large car datasets.

The rapid development of generative adversarial networks has significantly advanced the generation of synthetic images, presenting valuable opportunities and ethical dilemmas in their potential misuse across various industries. The necessity to distinguish real from AI-generated content is becoming increasingly critical to preserve the integrity of online data. While traditional methods for detecting fake images resulting from image tampering rely on hand-crafted features, the sophistication of manipulated images produced by generative adversarial networks requires more advanced detection approaches. The lightweight approach proposed here is based on convolutional neural networks that comprise only eight convolutional and two hidden layers that effectively differentiate AI-generated images from real ones. The proposed approach was assessed using two benchmark datasets and custom-generated data from Sentinel-2 imagery. It demonstrated superior performance compared to four state-of-the-art methods on the CIFAKE dataset, achieving the highest accuracy of 97.32%, on par with the highest-performing state-of-the-art method. Explainable AI is utilized to enhance our comprehension of the complex processes involved in synthetic image recognition. We have shown that, unlike authentic images, where activations often center around the main object, in synthetic images, activations cluster around the edges of objects, in the background, or in areas with complex textures.

Multidisciplinary approaches in science are still rare, especially in completely different fields such as agronomy science and computer science. We aim to create a state-of-the-art floating ebb and flow system greenhouse that can be used in future scientific experiments. The objective is to create a self-sufficient greenhouse with sensors, cloud connectivity, and artificial intelligence for real-time data processing and decision making. We investigated various approaches and proposed an optimal solution that can be used in much future research on plant growth in floating ebb and flow systems. A novel microclimate pocket-detection solution is proposed using an automatically guided suspended platform sensor system. Furthermore, we propose a methodology for replacing sensor data knowledge with artificial intelligence for plant health estimation. Plant health estimation allows longer ebb periods and increases the nutrient level in the final product. With intelligent design and the use of artificial intelligence algorithms, we will reduce the cost of plant research and increase the usability and reliability of research data. Thus, our newly developed greenhouse would be more suitable for plant growth research and production.

This paper presents a solution for upgrading a previous device model to an Industry 4.0 smart device, with the goal of maintaining high compatibility. A novel IoT architecture is presented that satisfies the characteristics of a smart device. We analysed existing IoT architectures and proposed a new architecture to achieve long-term security and usability. To ensure long-term security, we eliminated the possibility of device configuration outside the immediate vicinity of the device with a dedicated protocol. The security concepts of the existing architectures were also analysed and further modified. To improve compatibility with previous device models, we propose a new method to collect data from sensors by introducing a multithreaded microcontroller. We propose additional software components to ensure factory programming, maintenance, and cloud Big Data analysis. Based on our experiments, we adapted the algorithm to increase the accuracy of the temperature and flow sensors by using a temperature calibration device and known flow cycles. Measurement results are presented to confirm the successful upgrade. We designed a hardware architecture to ensure compatibility with previous and future device models. Issues with previous sensors encountered during the upgrade were discussed and resolved. A novel software architecture based on security for long-term IoT devices is proposed.