Wyniki wyszukiwania - BazTech

1

Learning-free deep features for multispectral palm-print classification

Aounallah Asma, Meraoumia Abdallah, Bendjenna Hakim

Computer Science

|

2023

|

T. 24 (2)

243--271

EN

The feature-extraction step is a major and crucial step in analyzing and understanding raw data, as it has a considerable impact on system accuracy. Despite the very acceptable results that have been obtained by many handcrafted methods, these can unfortunately have difficulty representing features in the cases of large databases or with strongly correlated samples. In this context, we attempt to examine the discriminability of texture features by proposing a novel, simple, and lightweight method for deep feature extraction to characterize the discriminative power of different textures. We evaluated the performance of our method by using a palm print-based biometric system, and the experimental results (using the CASIA multispectral palm--print database) demonstrate the superiority of the proposed method over the latest handcrafted and deep methods.

2

Fast multispectral deep fusion networks

Osin V., Cichocki A., Burnaev E.

Bulletin of the Polish Academy of Sciences. Technical Sciences

|

2018

|

Vol. 66, nr 6

875--889

EN

Most current state-of-the-art computer vision algorithms use images captured by cameras, which operate in the visible spectral range as input data. Thus, image recognition systems that build on top of those algorithms can not provide acceptable recognition quality in poor lighting conditions, e.g. during nighttime. Another significant limitation of such systems is high demand for computational resources, which makes them impossible to use on low-powered embedded systems without GPU support. This work attempts to create an algorithm for pattern recognition that will consolidate data from visible and infrared spectral ranges and allow near real-time performance on embedded systems with infrared and visible sensors. First, we analyze existing methods of combining data from different spectral ranges for object detection task. Based on the analysis, an architecture of a deep convolutional neural network is proposed for the fusion of multi-spectral data. This architecture is based on the single shot multi-box detection algorithm. Comparison analysis of the proposed architecture with previously proposed solutions for the multi-spectral object detection task shows comparable or better detection accuracy with previous algorithms and significant improvement of the running time on embedded systems. This study was conducted in collaboration with Philips Lighting Research Lab and solutions based on the proposed architecture will be used in image recognition systems for the next generation of intelligent lighting systems. Thus, the main scientific outcomes of this work include an algorithm for multi-spectral pattern recognition based on convolutional neural networks, as well as a modification of detection algorithms for working on embedded systems.

3

Supervised and unsupervised segmentation of multispectral retina images

Świtoński A., Błachowicz T., Misiuk-Hojło M., Wojciechowski K.

Przegląd Elektrotechniczny

|

2012

|

R. 88, nr 12b

111-114

EN

The segmentation method of multispectral human eye images suitable in ophthalmic diagnosis of structural retinal features characteristic for glaucoma and diabetic retinopathy diseases is presented. A multispectral imaging was realized in 21 spectral windows, between 400nm and 740nm, on a base of liquid crystal tunable filter and a high sensitivity monochrome camera. Results of supervised and unsupervised segmentation procedures of retina images, adopted from a color fundus device, are presented.

PL

W pracy zaprezentowano metodę segmentacji wielospektralnych obrazów dna oka ukierunkowaną na diagnostykę schorzeń jaskry i retinopatii cukrzycowej. Akwizycja wielospektralna prowadzona jest w 21 oknach widma z zakresu 400nm do 740nm na bazie elektronicznie sterowalnego filtra ciekłokrystalicznego i wysokiej czułości monochromatycznej kamery CCD. Przedstawiono uzyskane wyniki segmentacji w podejściu nadzorowanym i nienadzorowanym (Nadzorowana i nienadzorowana segmentacja wielospektralnych obrazów dna oka).

4

A computer-based imaging system for multispectral inspection of skin cancer

Świtoński A., Błachowicz T., Sieroń A., Wojciechowski K.

Przegląd Elektrotechniczny

|

2012

|

R. 88, nr 12b

107-110

EN

Multi-band imaging computer-based system, self-designed and self-constructed, based on a liquid-crystal filter with spectral transmittance driven in 400 nm - 740 nm wavelengths range is presented. Performed tests of images dimensionality reduction, which base on different types of principal component analysis, indicated onto flexibility and usefulness of the described approach for skin cancer diagnosis.

PL

W artykule opisano samodzielnie zaprojektowany i wykonany, wspomagany komputerowo, oparty na transmisyjnym filtrze ciekłokrystalicznym pracującym w zakresie długości fal od 400 nm do 740 nm, układ do obrazowania wielospektralnego. Przeprowadzone testy redukcji wymiarowości obrazu, w oparciu o różne rodzaje analizy jego składowych głównych, wskazały na użyteczność zastosowanego podejścia w diagnostyce raka skóry. (Komputerowy system wielospektralnej analizy danych obrazowych raka skóry).

5

Lasers in the dual use technologies

Mierczyk Z.

Bulletin of the Polish Academy of Sciences. Technical Sciences

|

2012

|

Vol. 60, nr 4

691-696

EN

In the more developed countries the use of modern technologies in the national economy is a general process allowing joint funding of a security research and development by both sources, state and private. The last one is especially involved in applications of modern technologies. One of the examples of important modern technologies being under development in many countries are dual-use technologies, which include IT technologies, sensors, effective energy sources, material science, nanotechnology, micro- and nano-electronics, photonics, biotechnology and quantum medicine. In this paper chosen technologies fulfilling the needs of the military technique and security monitoring systems, which have found their applications in the different branches of industry like power engineering, transportation, construction industry, metrology, protection of environment and the medicine, are discussed. The examples include the devices and lasers systems for different threats monitoring, which have been developed at the Military University of Technology. The research studies carried out on the analysis of various materials based on their spectroscopic characteristics: absorption, emission, dispersion, polarization and fluorescence in different mediums have led to the development of laser telemetry devices, environment monitoring devices and spatial imagery, as well as devices for medical diagnostics and therapy. Mentioned systems are composed of functional modules, which were developed to meet the real needs. These systems can be expanded further by addition of extra detectors of chemical materials and physical properties, and improving measuring functions and data transmission and processing.

6

Klasyfikator jądrowej wersji maszyny wektorów podpierających wspomagający diagnostykę obszarów nowotworowych w wielospektralnym obrazowaniu endoskopowym

Bieda R., Świtoński A., Kwiatek S., Latos W., Cieślar G., Sieroń A., Wojciechowski K.

Przegląd Elektrotechniczny

|

2010

|

R. 86, nr 12

13-16

PL

W pracy przedstawiono podstawy teoretyczne klasyfikatora maszyny wektorów podpierających oraz uzyskane przy jego zastosowaniu wyniki klasyfikacji obrazów wielospektralnych pozyskanych z 21 kanałowego systemu wielospektralnego obrazowania endoskopowego. Uzyskana rozróżnialność obszarów zmienionych chorobowo jest w ocenie lekarzy diagnostów wyższa niż w przypadku systemu Xillix Onco LIFE.

EN

The paper presents the theoretical basis and results obtained with use of the support vector machine classifier on multispectral image classification gained with 21-channel system for endoscopic multispectral imaging. Obtained distinguishability of pathological changes areas is higher than in the case of Xillix Onco LIFE system in the medical diagnosticians opinion.

7

Object recognition using a multi-spectral imaging system

Tominaga S.

Machine Graphics and Vision

|

2002

|

Vol. 11, No. 2/3

221-240

EN

A spectral-imaging system and algorithms for identifying obects in a natural scene based on surface-spectral reflectances are described. The imaging system in composed of a liquid-crystals tunable filter, a monochrome CCD camera, and a personal computer. The tunable filter is convenient for spectral imaging because the wavelenght band can ce changed easily and electronically. It is described how we can recover the surface-spectral refle tances of natural objects by using the multi-spectral imaging sysems. Algorithms are presented for estimating both spectral functions of the illuminant spectral--power distribution and suface-spectral reflectance from the spectral image data. Moreover, effective image processing procedures are proposed for highlight extraction and region segmentation. The segmentation is based on a pixel classification method using only the maximum sensor outputs. The overall performance of the proposed imaging system and algorithms is examined in an experiment using natural products, in which 21 spectral images are acquired in the wavelenght range of 450-650 nm.