2013Proceedings of SPIE, the International Society for Optical Engineering/Proceedings of SPIERequires access

An algorithm used for quality criterion automatic measurement of band-pass filters and its device implementation

Qianshun Liu, Yan Liu, Feihong Yu

Open publisher page 0 citations

Abstract

As a kind of film device, band-pass filter is widely used in pattern recognition, infrared detection, optical fiber communication, etc. In this paper, an algorithm for automatic measurement of band-pass filter quality criterion is proposed based on the proven theory calculation of derivate spectral transmittance of filter formula. Firstly, wavelet transform to reduce spectrum data noises is used. Secondly, combining with the Gaussian curve fitting and least squares method, the algorithm fits spectrum curve and searches the peak. Finally, some parameters for judging band-pass filter quality are figure out. Based on the algorithm, a pipeline for band-pass filters automatic measurement system has been designed that can scan the filter array automatically and display spectral transmittance of each filter. At the same time, the system compares the measuring result with the user defined standards to determine if the filter is qualified or not. The qualified product will be market with green color, and the unqualified product will be marked with red color. With the experiments verification, the automatic measurement system basically realized comprehensive, accurate and rapid measurement of band-pass filter quality and achieved the expected results.

About this research paper

What this paper is about

As a kind of film device, band-pass filter is widely used in pattern recognition, infrared detection, optical fiber communication, etc. In this paper, an algorithm for automatic measurement of band-pass filter quality criterion is proposed based on the proven theory calculation of derivate spectral transmittance of filter formula. Firstly, wavelet transform to reduce spectrum data noises is used. Secondly, combining with the Gaussian curve fitting and least squares method, the algorithm fits spectrum curve and searches the peak. Finally, some parameters for judging band-pass filter quality are figure out. Based on the algorithm, a pipeline for band-pass filters automatic measurement system has been designed that can scan the filter array automatically and display spectral transmittance of each filter. At the same time, the system compares the measuring result with the user defined standards to determine if the filter is qualified or not. The qualified product will be market with green color, and the unqualified product will be marked with red color. With the experiments verification, the automatic measurement system basically realized comprehensive, accurate and rapid measurement of band-pass filter quality and achieved the expected results.

Why it matters

A significance statement is not available in the OpenAlex record.

Key contribution

A contribution statement is not available in the OpenAlex record.

Method / approach

Method details are not available in the OpenAlex metadata.

Main findings

Findings are not separately available in the OpenAlex metadata.

Limitations

Limitations are not available in the OpenAlex metadata.

Applications

Application details are not available in the OpenAlex metadata.

Available abstract

As a kind of film device, band-pass filter is widely used in pattern recognition, infrared detection, optical fiber communication, etc. In this paper, an algorithm for automatic measurement of band-pass filter quality criterion is proposed based on the proven theory calculation of derivate spectral transmittance of filter formula. Firstly, wavelet transform to reduce spectrum data noises is used. Secondly, combining with the Gaussian curve fitting and least squares method, the algorithm fits spectrum curve and searches the peak. Finally, some parameters for judging band-pass filter quality are figure out. Based on the algorithm, a pipeline for band-pass filters automatic measurement system has been designed that can scan the filter array automatically and display spectral transmittance of each filter. At the same time, the system compares the measuring result with the user defined standards to determine if the filter is qualified or not. The qualified product will be market with green color, and the unqualified product will be marked with red color. With the experiments verification, the automatic measurement system basically realized comprehensive, accurate and rapid measurement of band-pass filter quality and achieved the expected results.

Key concepts: Computer science, High-pass filter, Filter (signal processing), Algorithm, Filter design, Low-pass filter, Gaussian filter, Band-pass filter

Related papers

Back to paper searchBrowse research topicsOriginal source
An algorithm used for quality criterion automatic measurement of band-pass filters and its device implementation — Research Paper | ScholarLens