Moskva, Moscow, Russian Federation
Moskva, Moscow, Russian Federation
The article proposes the classification of lenses with different symmetrical beam angles and offers a scale as a spot-light’s palette. A collection of spotlight’s images was created and classified according to the proposed scale. The analysis of 788 pcs of existing lenses and reflectors with different LEDs and COBs carried out, and the dependence of the axial light intensity from beam angle was obtained. A transfer training of new deep convolutional neural network (CNN) based on the pre-trained GoogleNet was performed using this collection. GradCAM analysis showed that the trained network correctly identifies the features of objects. This work allows us to classify arbitrary spotlights with an accuracy of about 80 %. Thus, light designer can determine the class of spotlight and corresponding type of lens with its technical parameters using this new model based on CCN.
LED luminaires classification, deep convolutional neural networks, lighting design, psychophysics of perception
1. Fairchild M.D. Color Appearance Models. John Wiley & Sons, Ltd, 2013. – 450 p. – ISBN 9781118653128.
2. Ajzenberg, Ju.B. O novoj svetotehnicheskoj klassifikacii svetil'nikov / Ju.B. Ajzenberg, G.M. Knorring // Svetotehnika. – 1968. – № 11. – S. 1-4.
3. GOST R 54350-2015. Pribory osvetitel'nye. Svetotehnicheskie trebovanija i metody ispytanij: nacional'nyj standart rossijskoj federacii pribory osvetitel'nye. Svetotehnicheskie trebovanija i metody ispytanij // Tehjekspert: [sajt]. – URL: http://docs.cntd.ru/document/1200121088 (data obrashhenija: 06.11.2020).
4. Nikolenko, S.I. Glubokoe obuchenie. Pogruzhenie v mir nejronnyhsetej / S.I. Nikolenko, A.A. Kadurin, E.V. Arhangel'skaja. – SPb.: Piter. – 2018. – 480 s. ISBN 978-5-4461-1537-2.
5. Deep Learning Toolbox. Getting Started Guide. R2020b / H.B. Mark, T.H. Martin, B.D. Howard. – Tekst: jelektronnyj // MATLAB Documentation: [sajt]. – URL: https://www.mathworks.com/help/pdf_doc/deeplearning/n net_gs.pdf (data obrashhenija: 06.11.2020).
6. Deep Learning Toolbox. Users Guide. R2020b / H.B. Mark, T.H. Martin, B.D. Howard. – Tekst: jelektronnyj // MATLAB Documentation: [sajt]. – URL: https://www.mathworks.com/help/pdf_doc/deeplearning/n net_ug.pdf (data obrashhenija: 06.11.2020).
7. Deep Learning Toolbox. Reference. R2020b / H.B. Mark, T.H. Martin, B.D. Howard. – Tekst: jelektronnyj // MATLAB Documentation: [sajt]. – URL: https://www.mathworks.com/help/pdf_doc/deeplearning/n net_ref.pdf (data obrashhenija: 06.11.2020).
