The encoding of the spatial-temporal archeological, historical and anthropological records can be... more The encoding of the spatial-temporal archeological, historical and anthropological records can be considered an ideal-typical representation of the human reasoning and thus also an artificial membrane interposed between the researcher and the past. These membranes are here considered artificial networks and can undergo interrogation processes through the most advanced analytical tools for learning and modeling complex configurations. The aim of this paper is to synthesize recent advances in Artificial Intelligence and Computer Science and – at the same time – to support the connectionists and symbolic computational paradigms as a new epistemic frontier in the automatic annotation of tangible and intangible heritage as well in the contemporary theories and methods of the archeological thought.
In the last decades the photogrammetry has undergone interesting innovation, both in terms of dat... more In the last decades the photogrammetry has undergone interesting innovation, both in terms of data processing and acquisition mode, to allow obtaining detailed 3D models useful for complete survey and important support for the management and recovery of cultural heritage and buildings. However, despite recent developments, the main photogrammetry outputs are raster data (ortophoto and DEM) and point clouds characterized by high informative content, but they are not typically extracted automatically. Automated feature detection is yet manual, time-consuming procedure and an active area of research. The raster to vector conversion is not direct, but transformations must be performed on the input data to convert the pixel values into features. Always, segmentations are preceded by filter technique to remove noise and to improve the conversion phase. However, remote sensing data and especially UAV photogrammetry output are the most complex to treat because of their heterogeneity (presence of different objects and shapes), the nature of sensor used and the different scale. In this work we experiment new enhancement filter to improve the automatic extraction of vector information for a UAV photogrammetry results of the facing walls of eminent church, symbol of the city of L’Aquila, the” Basilica of Santa Maria di Collemaggio”.
The encoding of the spatial-temporal archeological, historical and anthropological records can be... more The encoding of the spatial-temporal archeological, historical and anthropological records can be considered an ideal-typical representation of the human reasoning and thus also an artificial membrane interposed between the researcher and the past. These membranes are here considered artificial networks and can undergo interrogation processes through the most advanced analytical tools for learning and modeling complex configurations. The aim of this paper is to synthesize recent advances in Artificial Intelligence and Computer Science and – at the same time – to support the connectionists and symbolic computational paradigms as a new epistemic frontier in the automatic annotation of tangible and intangible heritage as well in the contemporary theories and methods of the archeological thought.
In the last decades the photogrammetry has undergone interesting innovation, both in terms of dat... more In the last decades the photogrammetry has undergone interesting innovation, both in terms of data processing and acquisition mode, to allow obtaining detailed 3D models useful for complete survey and important support for the management and recovery of cultural heritage and buildings. However, despite recent developments, the main photogrammetry outputs are raster data (ortophoto and DEM) and point clouds characterized by high informative content, but they are not typically extracted automatically. Automated feature detection is yet manual, time-consuming procedure and an active area of research. The raster to vector conversion is not direct, but transformations must be performed on the input data to convert the pixel values into features. Always, segmentations are preceded by filter technique to remove noise and to improve the conversion phase. However, remote sensing data and especially UAV photogrammetry output are the most complex to treat because of their heterogeneity (presence of different objects and shapes), the nature of sensor used and the different scale. In this work we experiment new enhancement filter to improve the automatic extraction of vector information for a UAV photogrammetry results of the facing walls of eminent church, symbol of the city of L’Aquila, the” Basilica of Santa Maria di Collemaggio”.
In this chapter we present a new unsupervised artificial adaptive system, able to extract feature... more In this chapter we present a new unsupervised artificial adaptive system, able to extract features of interest in digital imaging, to reduce image noise maintaining the spatial resolution of high contrast structures and the expression of hidden morphological features. The new system, named J-Net, belongs to the family of ACM systems developed by Semeion Research Center. J-Net is able to isolate in an almost geological way different brightness layers in the same image. These layers seem to be invisible to the human eye and for the other mathematical imaging system. This ability of the J-Net can have important medical applications.
It has been shown that a new procedure (Implicit Function as Squashing Time, IFAST) based on Arti... more It has been shown that a new procedure (Implicit Function as Squashing Time, IFAST) based on Artificial Neural Networks (ANNs) is able to compress eyes-closed resting electroencephalographic (EEG) data into spatial invariants of the instant voltage distributions for an automatic classification of mild cognitive impairment (MCI) and Alzheimer's disease (AD) subjects with classification accuracy of individual subjects higher than 92%. In this chapter the method has been applied to distinguish individual normal elderly (Nold) vs. Mild Cognitive Impairment (MCI) subjects, an important issue for the screening of large populations at high risk of AD. Eyes-closed resting EEG data (10-20 electrode montage) were recorded in 171 Nold and in 115 amnesic MCI subjects. The data inputs for the classification by IFAST were the weights of the connections within a non linear auto-associative ANN trained to generate the instant voltage distributions of 60-s ar-tifact free EEG data. The most relevant features were selected and coincidently the dataset was split into two halves for the final binary classification (training and testing) performed by a supervised ANN. The classification of the individual Nold and MCI subjects reached 95.87% of sensitivity and 91.06% of specificity (93.46% of accuracy). These results indicate that IFAST can reliably distinguish eyes-closed resting EEG in individual Nold and MCI subjects, and may be used for large-scale periodic screening of large populations at risk of AD and personalized care.
On tasks such as the recognition of handwritten digits, traditional methods from machine learning... more On tasks such as the recognition of handwritten digits, traditional methods from machine learning and computer vision have always failed to beat human performance. Inspired by the importance of diversity in biological system, we built an heterogeneous system that could achieve this goal. Our architecture could be summarized in two basic steps. First, we generate a diverse set of classification hypothesis using both Convo-lutional Neural Networks, currently the state-of-the-art technique for this task, among with other traditional and innovative machine learning techniques. Then, we optimally combine them through Meta-Nets, a family of recently developed and performing ensemble methods. In the resulting parliament of classifiers all hypothesis, despite of their accuracy or methodology, are considered. On the very competitive MNIST handwriting benchmark, our method is the first to beat human performance with 0.17% error rate, surprisingly showing diversity to be the key for success in decision making.
Uploads
Papers by massimo buscema