Machine Learning
A non-speech dataset built by a nine-classification discrete emotion model can be used for emotion recognition. The nine emotion dimensions are happy, angry, fearful, sad, brave, favorite, neutral, and calm. Recognition is performed by extracting their 384-dimensional feature vectors, which contain F0 envelopes, short-time excess zero rates, twelve-dimensional Meier cepstral coefficients, etc., as well as their first-order low-pass data. A total of 2,337 data samples were available.
- Categories:
19 Views
This dataset contains the measurement in an ultrawide band (UWB) system in the 6.5 GHz band, considering the presence of the human body as the only obstacle. There are measurements in line-of-sight condition to compare the results of the analysis performed. The measurements are part of our research on the adverse effects of the body shadowing in pedestrian localization systems.
The UWB system has four fixed terminals and one mobile terminal. The mobile terminal is worn on the chest of the person.
- Categories:
192 Views
We introduce a high-performance computer vision based Intraveneous (IV) infusion speed measurement system as a camera application on an iPhone or Android phone. Our system uses You Only Look Once version 5 (YOLOv5) as it was designed for real-time object detection, making it substantially faster than two-stage algorithms such as R-CNN. In addition, YOLOv5 offers greater precision than its predecessors, making it more competitive with other object detection methods.
- Categories:
19 Views
Problems related to ventral hernia are very common, and evaluating them using computational methods can assist in selecting the most appropriate treatment. This study collected data from over 3500 patients from different European countries observed during last 11 years (2012-2022), which were collected by specialists in hernia surgery. The majority of patients underwent standard surgical procedures, with a growing trend towards robotic surgery. This paper focuses on statistically evaluating the treatment methods in relation to patient age, body mass index (BMI), and the type of repair.
- Categories:
233 Views
Indian Rice Disease dataset (IRDD) contains rice leaf images of two classes namely BrownSpot and Healthy. The images are taken under various lightning conditions. Some images contain dew drops on the leaves. The rice leaf images are gatherd from fields in West Bengal, India. These images have been taken using smartphone camera by the project team members of IIIT Kalyani and IIT Kharagpur. The images are annotated and verified by the domain experts. This datset is a part of the project entitled "AI for Agriculture and Food Sustainability" funded by MeitY, Govt of India.
- Categories:
338 Views
IMUs have gained popularity for tracking joint kinematics due to their portability and versatility. However, challenges such as limited accuracy, lack of real-time data analysis, and complex sensor-to-segment calibration procedures have hindered their widespread use. To address these limitations, we developed a portable system that integrates four IMUs to collect treadmill walking data, with ground truth values obtained from a Motion Capture System.
- Categories:
69 Views
This study presents a benchmark for evaluating action-constrained reinforcement learning (RL) algorithms. In action-constrained RL, each action taken by the learning system must comply with certain constraints. These constraints are crucial for ensuring the feasibility and safety of actions in real-world systems. We evaluate existing algorithms and their novel variants across multiple robotics control environments, encompassing multiple action constraint types.
- Categories:
91 Views
This dataset focuses on cooperative spectrum sensing in a cognitive radio network, where multiple secondary users collaborate to detect the presence of a primary user. We introduce multiple cooperative spectrum sensing schemes based on a tree deep neural network architecture, incorporating a one-dimensional convolutional neural network and a long short-term memory network. The primary objective of these schemes is to effectively learn the activity pattern of the primary user.
- Categories:
1275 Views
Text classification systems have become increasingly important in recent years due to the explosion of online documents and the need to sort them for specific services. One of the most critical issues in text classification is the limited availability and diversity of datasets, which can lead to overfitting and poor generalization. In this context, we present a new dataset named Global News 60K (GN60K), which consists of 60,000 news articles from different sources from different parts of the world, covering 10 topics.
- Categories:
226 Views
A long-standing problem in thermal imaging is the inherent assumption of a uniform and known emissivity across an entire image. Semantic segmentation of the materials in a thermal image can identify the pixel-wise emissivity, thus rectifying the spatially uniform emissivity assumption with no human intervention. We have created a multispectral thermal image dataset consisting of nine materials (acrylic, aluminum, bakelite, ceramic, cork, EVA, granite, maple, and silicone) at six different temperatures.
- Categories:
218 Views