The classification of the images taken during the measurement of an architectural asset is an essential task within the digital documentation of cultural heritage. A large number of images are usually handled, so their classification is a tedious task (and therefore prone to errors) and habitually consumes a lot of time. The availability of automatic techniques to facilitate these sorting tasks would improve an important part of the digital documentation process. In addition, a correct classification of the available images allows better management and more efficient searches through specific terms, thus helping in the tasks of studying and interpreting the heritage asset in question. The main objective of this article is the application of techniques based on deep learning for the classification of images of architectural heritage, specifically through the use of convolutional neural networks. For this, the utility of training these networks from scratch or only fine tuning pre-trained networks is evaluated. All this has been applied to classifying elements of interest in images of buildings with architectural heritage value. As no datasets of this type, suitable for network training, have been located, a new dataset has been created and made available to the public. Promising results have been obtained in terms of accuracy and it is considered that the application of these techniques can contribute significantly to the digital documentation of architectural heritage.

In this article, a system for the detection of cracks in concrete tunnel surfaces, based on image sensors, is presented. Both data acquisition and processing are covered. Linear cameras and proper lighting are used for data acquisition. The required resolution of the camera sensors and the number of cameras is discussed in terms of the crack size and the tunnel type. Data processing is done by applying a new method called Gabor filter invariant to rotation, allowing the detection of cracks in any direction. The parameter values of this filter are set by using a modified genetic algorithm based on the Differential Evolution optimization method. The detection of the pixels belonging to cracks is obtained to a balanced accuracy of 95.27%, thus improving the results of previous approaches.

The development of deep neural networks that has been carried out in recent years allows solving highly complex computer vision classification problems. Often, although the results obtained with these classifiers are high, there are certain sectors that seek greater accuracy from these systems. Increasing the accuracy of neural networks can be achieved through ensemble learning, which combines different classifiers with the aim of selecting a winner based on different criteria about them. These techniques have traditionally shown good results although they involve training models of different nature and can even produce an overfitting with respect to the training data, so datasets must be chosen to correctly evaluate the result. In this paper, a Cross-Validation-Voting (CVV) technique for grocery product classification is presented. This technique improves several single state-of-the-art classifiers without combining different ones and avoids the problems of overfitting with respect to the training set. The single classifiers are trained multiple times against distributed sets to show how the results obtained to date from the classification of a well-known dataset are improved. In this dataset, an extensive test set was previously selected by the authors to show comparable results with other papers in the literature. The technique is valid not only for vision nets and can be used to solve numerous problems with different kinds of neural networks and classifiers.

Over the last few years, several techniques have been developed with the aim of implementing one-shot learning, a concept that allows classifying images with only a single image per training category. Conceptually, these methods seek to reproduce certain behavior that humans have. People are able to recognize a person they have only seen once, but they are probably not able to do the same with certain animals, such as a monkey. This is because our brains have been trained for years with images of people but not so much of animals. Among the one-shot learning techniques, some of them have used data generation, such as Generative Adversarial Networks (GAN). Other techniques have been based on the matching of descriptors traditionally used for object detection. Finally, one of the most prominent techniques involves using Siamese neural networks. Siamese networks are usually implemented with two convolutional nets that share their weights. They receive two images as input and can detect whether they belong to the same category or not. In the field of grocery products, there has been a lot of research on the one-shot learning problem but not so much on the use of Siamese networks. In this paper, several classifiers are firstly evaluated to decide on a convolutional model to be used with the Siamese and to improve the baseline results obtained in the dataset used. Then, two existing techniques are integrated within the Siamese model: a convolutional net and a Local Maximal Occurrence (LOMO) descriptor. The latter was initially used for the re-identification of people although it has shown its effectiveness to improve the values of a traditional Siamese with only convolutional sisters. The whole network is trained on categories and responds to different categories, showing its strong capacity to deal with the problem of having only one image per category.

Recent advances in artificial intelligence have intensified efforts to improve quality management in steel manufacturing. In this paper, we present the development and results of a system that aims to learn from the decisions made by experts to anticipate the problems that affect the final quality of the product in the steel rolling process. The system integrates a series of modules, including event filtering, automatic expert knowledge extraction, and decision-making neural networks, developed in a phased approach. The experimental results, using a three-year historical dataset, suggest that our system can anticipate quality issues with an accuracy of approximately 80%, enabling proactive defect prevention and a reduction in production losses. This approach demonstrates the potential for industrial AI applications for predictive quality assurance, highlighting the technical foundations and potential for industrial applications.