Wyniki wyszukiwania - BazTech

Ograniczanie wyników

Znaleziono wyników: 2

Liczba wyników na stronie

Wyniki wyszukiwania

Wyszukiwano:
w słowach kluczowych: uczenie maszyny

Sortuj według:

Ogranicz wyniki do:

Vibration-based identification of engine valve clearance using a convolutional neural network

Tabaszewski Maciej, Szymański Grzegorz M., Nowakowski Tomasz

Archives of Transport

2022

Vol. 61, iss. 1

117--131

Contemporary operation-related requirements for combustion engines force the necessity of ongoing assessment of their in operation technical condition (e.g. marine engines). The engine efficiency and durability depend on a variety of parameters. One of them is valve clearance. As has been proven in the paper, the assessment of the valve clearance can be based on vibration signals, which is not a problem in terms of signal measurement and processing and is not invasive into the engine structure. The authors described the experimental research aiming at providing information necessary to develop and validate the proposed method. Active experiments were used with the task of valve clearance and registration of vibrations using a three-axis transducer placed on the engine cylinder head. The tests were carried out during various operating conditions of the engine set by 5 rotational speeds and 5 load conditions. In order to extract the training examples, fragments of the signal related to the closing of individual valves were divided into 11 shorter portions. From each of them, an effective value of the signal was determined. Obtained total 32054 training vectors for each valve related to 4 classes of valve clearance including very sensitive clearance above 0.8 mm associated with high dynamic interactions in cylinder head. In the paper, the authors propose to use a convolutional network CNN to assess the correct engine valve clearance. The obtained results were compared with other methods of machine learning (pattern recognition network, random forest). Finally, using CNN the valve clearance class identification error was less than 1% for the intake valve and less than 3.5% for the exhaust valve. Developed method replaces the existing standard methods based on FFT and STFT combined with regression calculation where approximation error is up to 10%. Such results are more useful for further studies related not only to classification, but also to the prediction of the valve clearance condition in real engine operations.

Wavelet based classification of skin lesion images

Surówka G., Merkwirth C., Żabińska-Płazak E., Graca A.

Bio-Algorithms and Med-Systems

2006

Vol. 2, no. 4

43--49

Visual examination of the early stages of the melanocytic skin cancer (melanoma) may often lead to a false diagnosis. Only the resection and then histologic examination of the lesion can fully detect malignant transformations of the skin. This is the reason why development of non-invasive methods for dermatological diagnosis, like dermatoscopy, is of key importance. We build a MLP-based binary classifier for discriminating melanoma from dysplastic nevus utilizing textural information contained in the skin lesion images taken in dermatoscopic examinations. Our analysis is based on the multiresolution wavelet-based decomposition of the images. Significant features of both classes are found by means of the Ridge regression models. Discriminating melanoma from dysplastic nevus with this method yields a sensitivity and specificity of 89.5% and 90%, respectively.

Wizualna ocena wczesnych stanów procesu nowotworzenia skóry może prowadzić do błędnej diagnozy. Jedynie resekcja oraz histologiczna ocena może ocenić obecność procesu nowotworzenia. Stąd potrzeba nieinwazyjnej oceny w dermatologii jest potrzebą chwili. Zbudowaliśmy bazujący na MLP binarny klasyfikator dla dyskryminacji melanoma w oparciu o obrazy uzyskane dermatoskopowo. Metoda bazuje na dekompozycji obrazu. Model regresji Ridge'go został zaadaptowany dla klasyfikacji obrazu co dało specyficzność oceny rzędu 89.5% i 90%.