The Signal & Images Laboratory (SI-Lab) is an interdisciplinary research group in computer vision, signal analysis, intelligent vision systems and multimedia data understanding. It is part of the Institute of Information Science and Technologies (ISTI) of the National Research Council of Italy (CNR). This report accounts for the research activities of the Signal and Images Laboratory of the Institute of Information Science and Technologies during the year 2021

Medical waste, i.e. waste produced during medical activities in hospitals, clinics and laboratories, represents hazardous waste whose management involves special care and high costs. However, this kind of waste contains a significant fraction of highly valued materials that can enter a circular economy process. To this end, in this paper, we propose a computer vision approach for assisting in the primary sorting of medical waste. The feasibility of our approach is demonstrated on a representative dataset we collected and made available to the community, with which we have trained a model that achieves 100% accuracy, and a new dataset on which the trained model exhibits good generalization.

In this study, an automatic system based on open AI architectures was developed and fed with an in-house built image dataset to recognize seven of the most widespread and hard-to-control weeds in wheat in the Mediterranean environment. A total of 10,810 images were collected from the post-emergence (S1 dataset) to the pre-flowering stage (S2 dataset). A selection of pictures available from online sources (S3, 825 images) was used as a final and further independent test of the proposed recognition tool. The AI tool in the ensemble configuration achieved 100% accuracy on the validation and test set both for S1 and S2, while for S3 an accuracy of approximately 70% was achieved for weed species in the post-emergence stage.

Ensembling is a very well-known strategy consisting in fusing several different models to achieve a new model for classification or regression tasks. Ensembling has been proven to provide superior performance in various contexts related to pattern recognition and artificial intelligence. The winners of public challenges in image analysis often adopt solutions based on ensembling. The idea of ensembling has also provided suggestions for introducing recent deep learning architectures with multiple layer connections that mimic ensembling approaches. However, the full potential offered by ensembling is not yet fully exploited. This paper aims to explore possible research directions and define new fusion approaches. Preliminary experimental tests show favorable results with an increment in accuracy regarding the number of operations needed in training and inference.

This special issue of PRIA is devoted to some scientific results and trends of the 25th International Conference on Pattern Recognition (Virtual, Milano, Italy, January 10–15, 2021). Two important events of ICPR-2020 are represented in this special issue: (1) The paper of Professor Ching Yee Suen (Centre for Pattern Recognition and Machine Intelligence, Department of Computer Science and Software Engineering, Concordia University, Montreal, QC, Canada)–the recent winner of IAPR very prestigious K.S. Fu Prize for a year of 2020. The paper based on his lecture “From handwriting to human personality and facial beauty” presented at the ICPR 2020; (2) Special issue “ICPR-2020 Workshop “Image Mining. Theory and Applications.” The analysis of the scientific contribution of IMTA-VII-2021 allows us to draw the following conclusions: (1) The construction of a unified mathematical theory of image analysis is still far from complete. (2) There is considerable interest in the development of new mathematical methods for analyzing and evaluating information presented in the form of images. (3) There is a tendency to expand the mathematical apparatus in the development of new methods of image analysis and recognition by involving in this process areas of mathematics that were not previously used in image analysis. (4) The gap between the capabilities of new mathematical methods of image analysis and recognition and their actual use in solving applied problems remains significant. (5) There is an excessive use of neural networks in solving applied problems of image analysis and image recognition, and quite often without proper justification and interpretation of the results. The special issue includes articles based on the workshop papers selected by the IMTA-VII-2021 Program Committee for publication in PRIA . The PRIA special issue “Scientific Resume of the 25th International Conference on Pattern Recognition” is prepared by the National Committee for Pattern Recognition and Image Analysis of the Russian Academy of Sciences, the IAPR member society, and by the IAPR Technical Committee no. 16 on Algebraic and Discrete Mathematical Techniques in Pattern Recognition and Image Analysis.

The inspection of power lines is the crucial task for the safe operation of power transmission: its components require regular checking to detect damages and faults that are caused by corrosion or any other environmental agents and mechanical stress. During recent years, the use of Unmanned Autonomous Vehicle (UAV) for environmental and industrial monitoring is constantly growing and the demand for fast and robust algorithms for the analysis of the data acquired by drones during the inspections has increased. In this work, we use UAV to acquire power transmission lines data and apply image processing to highlight expected faults. Our method is based on a fusion algorithm for the infrared and visible power lines images, which is invariant to large scale changes and illumination changes in the real operating environment. Hence, different algorithms from image processing are applied to visible and infrared thermal data, to track the power lines and to detect faults and anomalies. The method significantly identifies edges and hot spots from the set of frames with good accuracy. At the final stage we identify hot spots using thermal images. The paper concludes with the description of the current work, which has been carried out in a research project, namely SCIADRO.

Gold Nanoparticles (GNs) have been widely used during the past few years in various technical and biomedical applications. In particular, the resonance optical properties of nanometer sized particles have been employed to design biochips and biosensors used as analytical tools. The optical properties of non-functionalized GNs and core-gold nanoshells play a crucial role for the design of biosensors where gold surface is used as sensing component. Due to plasmonic coupling with electromagnetic fields, GNs exhibit excellent optical tunability at visible and near-infrared frequencies leading to sharp peaks in their spectral extinction. In this paper, we study how the optical properties of gold nanoparticles and core-gold nanoshells are changed as a function of different sizes, shapes, composition and biomolecular coating with characteristic shifts towards near-infrared region, 750-900nm. We show that the optical tunability in this region can be carefully checked for particle sizes falling in the range 100-150nm. Such tunability in the range 750-900 nm is particularly important in biosystems, because the high transmittivity properties characterizing many biological tissues. using by using a Scanning Near Field Microscopy, some preliminary experimental results on the NIR response of GNs, when absorbed by mouse

Metal nanoshells are a type of nanoparticle composed by a dielectric core and a metallic coating. These nanoparticles have stimulated interest due to their remarkable optical properties. In common with metal colloids, they show distinctive absorption peaks at specific wavelengths due to surface plasmon resonance. However, unlike bare metal colloids, the wavelengths at which resonance occurs can be tuned by changing the core radius and coating thickness. One basic application of such property is in medicine, where it is hoped that nanoshells with absorption peaks in the near-infrared can be attached to cancerous cells. In this paper, we study the changes of optical response in visible and near infrared wavelengths from single to randomly distributed clusters of nanoshells. The results were obtained using a novel formulation of Mie theory in evanescent wave conditions, with a finite-difference time-domain (FDTD) simulation and experimentally on BaTiO3-gold nanoshells using a scanning near-optical microscope. The results show that the optical signal of a randomly distributed cluster of nanoshells can be supplementary tuned with respect to the case of single nanoshell depending by the geometric configuration of the clusters.

We propose a method designed for processing acoustic and optical data producing information about the presence of manmade and archaeological objects lying on the seabed. This method statistically highlights this type of artifacts among surrounding environment, weighting properly the persistence of meaningful curves in a video sequence, or in a sonogram. To this aim, we made use of the ELSD algorithm, a parameterless method inspired by Gestalt principles which has proven to provide promising results.

Let (W,S) be a Coxeter system, S finite, and let GW be the associated Artin group. One has configuration spaces Y, Y-W, where G(W) = pi(1)(Y-W), and a natural W -covering fW : Y -> YW. The Schwarz genus g (f(W)) is a natural topological invariant to consider. In [DS00] it was computed for all finite-type Artin groups, with the exception of case A(n) (for which see [Vas92], [DPS04]). In this paper we generalize this result by computing the Schwarz genus for a class of Artin groups, which includes the affinne-type Artin groups. Let K = K(W, S) be the simplicial scheme of all subsets J subset of S such that the parabolic group W-J is finite. We introduce the class of groups for which dim(K) equals the homological dimension of K, and we show that g (f(W)) is always the maximum possible for such class of groups. For affine Artin groups, such maximum reduces to the rank of the group. In general, it is given by dim(X-W) + 1, where X-W subset of Y-W is a well- known CW-complex which has the same homotopy type as Y-W.

The primary purpose of this work consists in treating optical and acoustic signals in order to extract useful information for applications in underwater archaeology. Data are processed to assess the presence of geometrically regular elements, potentially indicating handmade objects lying on the seafloor. Geometrical elements are recognized by means of suitable algorithms and their statistical persistence in the data stream is employed as a descriptor. Multi-sensor data are processed by applying segmentation and classification procedures based on a geometrical pattern analysis, with the purpose of discerning different materials. We basically seek for meaningful features in the data in order to perform robust object recognition, also in case of unfavorable environmental conditions. Finally we define a unique data fusion model that can be exploited for exhaustive interpretation of the underwater scene.

The impact of oil pollutions on coastal environment, concerns both the economy and the quality of life. The increasing importance of petroleum products and its maritime transportation raised the concern on navigation safety and environmental protection, leading to a major interest in frameworks for remotely detecting oil spill at sea. While many of the approaches have been focused on large oil spills, smaller ones and operational discharges in regional area received fairly less consideration. In this work we present a framework where, in addition to classical remote sensing the information is enriched with data collected in situ thanks to static and mobile sensors and thus leveraging on innovative methods for data correlation and fusion. The proposed GIS infrastructure is an integrated and interoperable system based on advanced sensing capabilities from a variety of electronic sensors along with geo-positioning tools, yet suitable for local authorities and stakeholders.

The Thesaurus Project, funded by the Regione Toscana, combines humanistic and technological research aiming at developing a new generation of cooperating Autonomous Underwater Vehicles and at documenting ancient and modern Tuscany shipwrecks. Technological research will allow performing an archaeological exploration mission through the use of a swarm of autonomous, smart and self-organizing underwater vehicles. Using acoustic communications, these vehicles will be able to exchange each other data related to the state of the exploration and then to adapt their behavior to improve the survey. The archival research and archaeological survey aim at collecting all reports related to the underwater evidences and the events of sinking occurred in the sea of Tuscany. The collected data will be organized in a specific database suitably modeled.

The wide availability of embedded sensor platforms and low-cost cameras—together with the developments in wireless communication—make it now possible the conception of pervasive intelligent systems based on vision. Such systems may be understood as distributed and collaborative sensor networks, able to produce, aggregate and process images in order to understand the observed scene and communicate the relevant information found about it. In this paper, we investigate the peculiarities of visual sensor networks with respect to standard vision systems and we identify possible strategies to accomplish image processing and analysis tasks over them. Although the rather strong constraints in computational and transmission power of embedded platforms that may prevent the use of state of the art computer vision and pattern recognition methods, we argue that multi-node processing methods may be envisaged to decompose a complex task into a hierarchy of computationally simpler problems to be solved over the nodes of the network. These ideas are illustrated by describing an application of visual sensor network to infomobility. In particular, we consider an experimental setting in which several views of a parking lot are acquired by the sensor nodes in the network. By integrating the various views, the network is capable to provide a description of the scene in terms of the available spaces in the parking lot.

We prove that the complement to the affine complex arrangement of type (B) over tilde (n) is a K(pi, 1) space. We also compute the cohomology of the affine Artin group G (B) over tilde (n) ( of type (B) over tilde (n)) with coefficients in interesting local systems. In particular, we consider the module Q [q+/-1; t+/-1]; where the first n standard generators of G (B) over tilde (n) act by (-q)-multiplication while the last generator acts by (-t)-multiplication. Such a representation generalizes the analogous 1-parameter representation related to the bundle structure over the complement to the discriminant hypersurface, endowed with the monodromy action of the associated Milnor fibre. The cohomology of G (B) over tilde (n) with trivial coefficients is derived from the previous one.

Progress is being made in the development of microanalytical systems for biosensing. Because the sensor signal-to-noise ratio increases with decreasing size for many devices, considerable effort to fabricate small sensors is going to be addressed. Due to their hollow cylindrical structure, carbon nanotubes (CNTs) are considered very promising for many potential nano-device applications. Fluorescence microscopy in the near-infrared (NIR) between 950 and 1600 nm has been developed as a novel method to image and study single-walled carbon nanotubes (SWNTs) in a variety of environments. Recently, hybridisation of DNA using NIR band-gap fluorescence has been experimentally demonstrated. We describe a numerical simulation, where the fluorescence shift energy is connected to exciton density variation when the molecular recognition is located on the SWNT immersed in a physiological solution.

This paper is devoted to the presentation of a framework for the description of anatomical structures, based both on topological and geometrical features and on semantic annotation. We argue that a 3D model-representing an anatomical structure—may be enhanced with other non-geometrical pieces of information relevant to the particular problem-context. Hybrid methods for similarity searches are then introduced and shown to be able to support effective case-based reasoning procedures. The approach is illustrated with examples from several medical application fields in order to discuss its potential impact.

In this paper, we present an approach to the description of time-varying anatomical structures. The main goal is to compactly but faithfully describe the whole heart cycle in such a way to allow for deformation pattern characterization and assessment. Using such an encoding, a reference database can be built, thus permitting similarity searches or data mining procedures.