Wyniki wyszukiwania - BazTech

1

Klasyfikacja akcji nóg w szermierce z użyciem akcelerometru

Malawski F.

Elektronika : konstrukcje, technologie, zastosowania

|

2017

|

Vol. 58, nr 4

8--12

PL

Automatyczna analiza oraz klasyfikacja akcji w sporcie stanowi użyteczne narzędzie dla zawodników oraz trenerów. W niniejszej pracy zaprezentowano efektywną metodę klasyfikacji akcji nóg w szermierce z wykorzystaniem akcelerometru. Do opisu ruchu wykorzystano zarówno cechy czasowe jak i częstotliwościowe. Do wyboru podzbioru cech o największej sile dyskryminacyjnej zastosowano metody selekcji cech oparte o algorytmy AdaBoost oraz Lasso. Klasyfikację akcji zrealizowano w oparciu o metody DTW oraz SVM. Przeprowadzono badania eksperymentalne na dedykowanej bazie danych z akcjami nóg w szermierce, która zawiera trudne do klasyfikacji akcje o podobnej trajektorii, ale różnej dynamice ruchu. Metody zaproponowane w niniejszej pracy umożliwiają uzyskanie 70% skuteczności klasyfikacji.

EN

Automatic analysis and classification of actions in sports constitutes a useful tool for both athletes and coaches. In this paper we present an efficient method for classification of actions in fencing footwork by employing an accelerometer. Fencers’ motion is described by both time and frequency domain features. In order to select a relevant subset of features we employ feature selection methods, namely AdaBoost and Lasso. In the classification we use DTW and SVM algorithms. Experiments were conducted on a dedicated dataset with fencing footwork actions, which contains actions, which are difficult to classify, as they have similar trajectories and vary mostly in the dynamics of the motion. The proposed methods achieved recognition accuracy better than 70%.

2

Compression of image sequences in interactive medical teleconsultations

Malawski F., Czekierda Ł.

Computer Science

|

2017

|

Vol. 18 (1)

95--114

EN

Interactive medical teleconsultations are an important tool in modern medical practice. Their applications include remote diagnostics, conferences, work- shops, and classes for students. In many cases, standard medium or low-end machines are employed, and the teleconsultation systems must be able to provide a high quality of user experience with very limited resources. Particularly problematic are large datasets consisting of image sequences that need to be accessed fluently. The main issue is insufficient internal memory; therefore, proper compression methods are crucial. However, a scenario where image sequences are kept in a compressed format in the internal memory and decompressed on- the-fly when displayed is difficult to implement due to performance issues. In this paper, we present methods for both lossy and lossless compression of medical image sequences that only require compatibility with the Pixel Shader 2.0 standard, which is present even on relatively old, low-end devices. Based on the evaluation of quality, size reduction, and performance, these methods have been proven to be suitable and beneficial for medical teleconsultation applications

3

Memory management in interactive medical teleconsultation systems

Malawski F.

Challenges of Modern Technology

|

2017

|

Vol. 8, no. 1

3--7

EN

Medical teleconsultation systems are an important tool in modern medicine, as they enable convenient, real-time collaboration of doctors from remote hospitals, without the need to travel. Due to the large size, efficient handling of medical imaging data in real-time teleconsultations is challenging. Memory management provided by operating systems is based on the virtual memory, which introduces considerable delays. In this paper we propose dedicated methods for memory management, including both memory monitoring and pre-loading of images. Results of the conducted experiments indicate, that the proposed methods can significantly improve the user experience.

4

Acquisition of databases for facial analysis

Malawski F.

Challenges of Modern Technology

|

2016

|

Vol. 7, no. 3

3--7

EN

This article describes guidelines and recommendations for acquisition of databases for facial analysis. New devices and methods for both face recognition and facial expression recognition are constantly developed. In order to evaluate these devices and methods, dedicated datasets are recorded. Acquisition of a database for facial analysis is not an easy task and requires taking into account multiple issues. Based on our experience with recording databases for facial expression recognition, we provide guidelines regarding the acquisition process. Multiple aspects of such process are discussed in this work, namely selection of sensors and data streams, design and structure of the database, technical aspects, acquisition conditions and design of the user interface. Recommendations how to address these aspects are provided and justified. An acquisition software, designed according to these guidelines, is also discussed. The software was used for recording an extended version of our previous facial expression recognition database and proved to both ensure correct data and be convenient for the recorded subjects.

5

Hand motion tracking with tablet devices as a human-computer interface for people with cerebral palsy

Barczewska K., Malawski F.

Challenges of Modern Technology

|

2015

|

Vol. 6, no. 2

3--9

EN

This article describes experiments which were conducted with two groups of people: healthy and with cerebral palsy. The purpose of the experiments was to verify two hypothesis: 1) that people with cerebral palsy can interact with computer using their natural gestures, 2) that games for tablets can be a research tool in experiments with young disabled people. Two games, which require the user to make simple gestures to accomplish given tasks, were designed and implemented on a tablet device with a built-in camera. Because of the camera limitations, the tracking process employs blue markers. By moving hand with a blue marker it was possible to perform navigation tasks in both games. In the first game the user had to gather, in 30 seconds, as many objects as possible. The objects were placed on the screen in a grid pattern. In the second game the user had to catch one object 20 times. The position of the object changed after each catch. Results obtained by healthy people were considered as a reference. However there are significant differences between measured parameters in both groups, all persons - healthy and with cerebral palsy - were able to accomplish the tasks by using simple gestures. Games for tablets turned out to be a very attractive research tool from the perspective of young, disabled users. They participated in measurement sessions much more willingly than in experiments without games.

6

Top-view people counting in public transportation using Kinect

Malawski F.

Challenges of Modern Technology

|

2014

|

Vol. 5, no. 4

17--20

EN

This article describes a method for people counting in public transportation. In this particular scenario, various body poses corresponding to holding handrails must be accounted for. Kinect sensor mounted vertically has been employed to acquire a database of images of 1-5 persons, with and without body poses of holding a handrail. An algorithm has been devised for robust people counting, consisting of multiple steps. The handrails are removed by substituting an average image of the handrails from the image with persons holding a handrail. The image is then processed in blocks in order to find potential local maxima, which are subsequently verified to find head candidates. Finally, non-head objects are filtered out, based on the ratio of pixels with similar and near-zero value, in the neighbourhood of the maxima. The method has an average accuracy of 91% and has proved to handle well the handrails in the depth maps.

7

Applying hand gesture recognition with time-of-flight camera for 3D medical data analysis

Malawski F.

Challenges of Modern Technology

|

2014

|

Vol. 5, no. 4

12--16

EN

This paper describes a human-computer interface based on hand gesture recognition, intended for analysis of 3D medical data. The gestures are designed to minimize the required muscle tension when using the system. Gesture recognition is based on a 3D sensor. Depth maps are acquired by a time-of-flight camera, designed specifically for hand gestures recognition. The depth images are denoised and segmented to right and left hand. The contours of the hands are found and a modified Shape Context descriptor is utilized for each hand, providing a set of features, which are employed to train and test various classifiers. Naive Bayes, Random Forest and Support Vector Machine (SVM) classifiers are utilized, with search of optimal parameters using cross-validation. The best accuracy (95%) is achieved with the Support Vector Machine classifier. The gestures are mapped to various controls of a 3D medical visualization module. Two visualization methods are employed - isosurface and cut-planes. The left hand is assigned to switching between different control modes and the right hand gestures are corresponding to controlling various properties in each mode. The system is convenient to use and runs in real-time on a typical PC machine.