9,081 research outputs found
CheeseMaking-IDB
No full textCheeseMaking-IDB is the public image dataset described in the following article: TBD# CM-IDB: The Cheese Making Image Database for Image Processing and AnalysisCM-IDB is the public image dataset described in the following article: TBDAuthors: Andrea Loddo, Cecilia Di Ruberto, Giuliano Armano, Andrea Manconi.# Dataset acquisitionThe dataset was obtained through the collaboration between Biosabbey s.r.l. and the Department of Mathematics and Computer Science (DMI) ofthe University of Cagliari under an agreement facilitated by DMI. The acquisition process was overseen by Massimiliano Sicilia of Laore Sardegna as part of his collaboration with Biosabbey s.r.l. Subsequently, Andrea Loddo, affiliated with DMI at the University of Cagliari, curated and organized the dataset, which is currently maintained by him.# NOTE: please indicate the following in case of using this dataset in your own work:TBD---------------------------------------------------------------------------------------------------------------# Description: The dataset for this study was assembled by collecting a series of 8 image sets from a Sardinian (Italy) dairy company (Podda Formaggi).Each set illustrates the coagulation process of milk, showing the transition from a liquid state to a gelatinous form known as curd. An image at different positions within the sequence for each set identifies the precise moment when this transformation begins, referred to as the curd-firming time.The images were taken using a camera with a CMOS sensor of size 35.9x24.0 mm and a resolution of 24 Mpixel, specifically a Nikon D750. All images are in RGB format, with a resolution of 6016x4016 pixels, and were taken at approximately 10-second intervals.Within each set, the images are organized chronologically and labeled based on the stage of maturation: pre-CF-time (negative class) or CF-time (positive class). The time interval between consecutive images varies, with negative examples taken every 10 seconds and positive examples taken every 2 seconds.Copyright (c) 2024 andrealoddo</p
EtherAddressLookup dataset
No full textDataset of scams gathered from https://github.com/409H/EtherAddressLooku
A novel deep learning based approach for seed image classification and retrieval
No full textSeeds image analysis has become essential to preserve biodiversity. This is why recognition and classification of plant species on the earth’s planet is nowadays a great challenge. The paper focuses on this purpose by studying two plant seeds datasets to classify their families or species through deep learning techniques. SeedNet, a novel CNN has been proposed to face the depicted issue, and several state-of-the-art convolutional neural networks have been exploited for an exhaustive comparison of most adequate for the considered scenario. In detail, promising results in seed classification for both analysed datasets, reaching accuracy values of 95.65% for the first one and 97.47% for the second one, have been obtained. The retrieval problem with the deep learning approach was also addressed, achieving satisfying performances. We consider the obtained results for both the tasks as an excellent starting point to develop a complete seeds recognition, classification and retrieval system to offer impressive support in agriculture and botany fields
On the Reliability of CNNs in Clinical Practice: A Computer-Aided Diagnosis System Case Study
No full textLeukocytes classification is essential to assess their number and status since they are the body’s first defence against infection and disease. Automation of the process can reduce the laborious manual process of review and diagnosis by operators and has been the subject of study for at least two decades. Most computer-aided systems exploit convolutional neural networks for classification purposes without any intermediate step to produce an accurate classification. This work explores the current limitations of deep learning-based methods applied to medical blood smear data. In particular, we consider leukocyte analysis oriented towards leukaemia prediction as a case study. In particular, we aim to demonstrate that a single classification step can undoubtedly lead to incorrect predictions or, worse, to correct predictions obtained with wrong indicators provided by the images. By generating new synthetic leukocyte data, it is possible to demonstrate that the inclusion of a fine-grained method, such as detection or segmentation, before classification is essential to allow the network to understand the adequate information on individual white blood cells correctly. The effectiveness of this study is thoroughly analysed and quantified through a series of experiments on a public data set of blood smears taken under a microscope. Experimental results show that residual networks perform statistically better in this scenario, even though they make correct predictions with incorrect information
On the Effectiveness of Leukocytes Classification Methods in a Real Application Scenario
No full textAutomating the analysis of digital microscopic images to identify the cell sub-types or the presence of illness has assumed a great importance since it aids the laborious manual process of review and diagnosis. In this paper, we have focused on the analysis of white blood cells. They are the body’s main defence against infections and diseases and, therefore, their reliable classification is very important. Current systems for leukocyte analysis are mainly dedicated to: counting, sub-types classification, disease detection or classification. Although these tasks seem very different, they share many steps in the analysis process, especially those dedicated to the detection of cells in blood smears. A very accurate detection step gives accurate results in the classification of white blood cells. Conversely, when detection is not accurate, it can adversely affect classification performance. However, it is very common in real-world applications that work on inaccurate or non-accurate regions. Many problems can affect detection results. They can be related to the quality of the blood smear images, e.g., colour and lighting conditions, absence of standards, or even density and presence of overlapping cells. To this end, we performed an in-depth investigation of the above scenario, simulating the regions produced by detection-based systems. We exploit various image descriptors combined with different classifiers, including CNNs, in order to evaluate which is the most suitable in such a scenario, when performing two different tasks: Classification of WBC subtypes and Leukaemia detection. Experimental results have shown that Convolutional Neural Networks are very robust in such a scenario, outperforming common machine learning techniques combined with hand-crafted descriptors. However, when exploiting appropriate images for model training, even simpler approaches can lead to accurate results in both tasks
Two dimensions of the intersemiotic legal translation in digital environments. Programming legal acts as a resemiotization process
No full textAn intersemiotic legal translation involves at least two different semiotic codes, of which at least one is not verbal.
The authors will draw a distinction between two macro dimensions of legal translation in digital environments. The first can be schematized in one direction from the machine to
the user. Intersemiotic translation is applied in the creation of interfaces (creation of buttons, forms, digital procedures) that are conditions of possibility for the fulfillment of certain legal acts (the acceptance of a user agreement, the purchase of a material or digital asset, the request for a particular document to the PA). The second dimension can be schematized in a direction that goes from the user to the machine. In this case, the user plays the role of the translator since he transposes, through the frontend, his contractual intent from natural language to machine language, using tools available to him by the developer. For example, a team of developers can exploit a Content Management Systems (CMS), which exposes two main interfaces, namely those that connect those who develop the interfaces and those who use them. Respectively, a backend that allows configuring the operating environment (user to machine) and, a frontend that exposes the communication interfaces with the end user (machine to user)
An intersemiotic translation of normative utterances to machine language
No full textProgramming Languages (PL) effectively performs an intersemiotic translation from a natural language to machine language. PL comprises a set of instructions to implement algorithms, i.e., to perform (computational) tasks. Similarly to Normative Languages (NoL), PLs are formal languages that can perform both regulative and constitutive
functions. The paper presents the first results of interdisciplinary research aimed at highlighting the similarities between NoL (social sciences) and PL (computer science)
through everyday life examples, exploiting Object-Oriented Programming Language tools and an Internet of Things (IoT) system as a case study. Given the pandemic emergency, the urge to move part of our social life to the digital world arose, together with the need to effectively transpose regulative rules and constitutive rules through different strategies for translating a normative utterance expressed in natural language
Soils of the Plio-Pleistocenic volcanic district of Montiferru (central-western Sardinia, Italy)
No full textA deep architecture based on attention mechanisms for effective end-to-end detection of early and mature malaria parasites in a realistic scenario
No full textBackground: Malaria is a critical and potentially fatal disease caused by the Plasmodium parasite and is responsible for more than 600,000 deaths globally. Early and accurate detection of malaria parasites is crucial for effective treatment, yet conventional microscopy faces limitations in variability and efficiency. Methods: We propose a novel computer-aided detection framework based on deep learning and attention mechanisms, extending the YOLO-SPAM and YOLO-PAM models. Our approach facilitates the detection and classification of malaria parasites across all infection stages and supports multi-species identification. Results: The framework was evaluated on three publicly available datasets, demonstrating high accuracy in detecting four distinct malaria species and their life stages. Comparative analysis against state-of-the-art methodologies indicates significant improvements in both detection rates and diagnostic utility. Conclusion: This study presents a robust solution for automated malaria detection, offering valuable support for pathologists and enhancing diagnostic practices in real-world scenarios
- …
