Material Segmentation Using 1-D Convolutional Neural Network With Transient Histogram

This study introduces a method for material classification using transient histograms obtained via a single-photon avalanche diode (SPAD) sensor. Temporal resolution in optical sensing plays a crucial role in material classification and surface segmentation, particularly for distinguishing materials...

Full description

Saved in:
Bibliographic Details
Main Authors: Yohanssen Pratama, Kazuya Kitano, Takahiro Kushida, Yuki Fujimura, Takuya Funatomi, Yasuhiro Mukaigawa
Format: Article
Language:English
Published: IEEE 2025-01-01
Series:IEEE Access
Subjects:
Online Access:https://ieeexplore.ieee.org/document/10942358/
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849702074067976192
author Yohanssen Pratama
Kazuya Kitano
Takahiro Kushida
Yuki Fujimura
Takuya Funatomi
Yasuhiro Mukaigawa
author_facet Yohanssen Pratama
Kazuya Kitano
Takahiro Kushida
Yuki Fujimura
Takuya Funatomi
Yasuhiro Mukaigawa
author_sort Yohanssen Pratama
collection DOAJ
description This study introduces a method for material classification using transient histograms obtained via a single-photon avalanche diode (SPAD) sensor. Temporal resolution in optical sensing plays a crucial role in material classification and surface segmentation, particularly for distinguishing materials with similar visual properties. In this study, SPAD sensors were utilized to capture transient histograms with temporal resolutions ranging from 13 picoseconds to 208 picoseconds, enabling precise extraction of temporal signatures for various materials. A comparative evaluation of classification techniques, including one-dimensional convolutional neural networks (1-D CNN), random forest (RF), support vector classifier (SVC), and k-nearest neighbors (KNN), was conducted to assess the impact of temporal resolution and exposure time on classification accuracy. 1-D CNN achieved the highest classification accuracy of 99.25% at a temporal resolution of 13 ps and an exposure time of 0.09 s, significantly outperforming other methods. Additionally, the proposed SPAD-based system was evaluated for material segmentation on non-planar surfaces. In a real-world experiment, 1-D CNN achieved an overall accuracy of 87.5% in differentiating visually similar materials, demonstrating the effectiveness of transient histograms for material classification where conventional RGB-based methods fail. These findings highlight the potential of SPAD sensors combined with advanced classification techniques to enhance material classification and segmentation, providing a versatile framework for applications in robotics, computer vision, and optical sensing.
format Article
id doaj-art-c29ad52d8b2b4772b607a5bbda5f6eea
institution DOAJ
issn 2169-3536
language English
publishDate 2025-01-01
publisher IEEE
record_format Article
series IEEE Access
spelling doaj-art-c29ad52d8b2b4772b607a5bbda5f6eea2025-08-20T03:17:46ZengIEEEIEEE Access2169-35362025-01-0113592955930810.1109/ACCESS.2025.355477610942358Material Segmentation Using 1-D Convolutional Neural Network With Transient HistogramYohanssen Pratama0https://orcid.org/0000-0001-6073-9016Kazuya Kitano1https://orcid.org/0000-0001-6430-6963Takahiro Kushida2https://orcid.org/0000-0002-6965-9781Yuki Fujimura3Takuya Funatomi4https://orcid.org/0000-0001-5588-5932Yasuhiro Mukaigawa5https://orcid.org/0000-0001-8689-3724Graduate School of Science and Technology, Nara Institute of Science and Technology, Ikoma, Nara, JapanGraduate School of Science and Technology, Nara Institute of Science and Technology, Ikoma, Nara, JapanCollege of Information Science and Engineering, Ritsumeikan University, Osaka, Ibaraki, JapanGraduate School of Science and Technology, Nara Institute of Science and Technology, Ikoma, Nara, JapanGraduate School of Science and Technology, Nara Institute of Science and Technology, Ikoma, Nara, JapanGraduate School of Science and Technology, Nara Institute of Science and Technology, Ikoma, Nara, JapanThis study introduces a method for material classification using transient histograms obtained via a single-photon avalanche diode (SPAD) sensor. Temporal resolution in optical sensing plays a crucial role in material classification and surface segmentation, particularly for distinguishing materials with similar visual properties. In this study, SPAD sensors were utilized to capture transient histograms with temporal resolutions ranging from 13 picoseconds to 208 picoseconds, enabling precise extraction of temporal signatures for various materials. A comparative evaluation of classification techniques, including one-dimensional convolutional neural networks (1-D CNN), random forest (RF), support vector classifier (SVC), and k-nearest neighbors (KNN), was conducted to assess the impact of temporal resolution and exposure time on classification accuracy. 1-D CNN achieved the highest classification accuracy of 99.25% at a temporal resolution of 13 ps and an exposure time of 0.09 s, significantly outperforming other methods. Additionally, the proposed SPAD-based system was evaluated for material segmentation on non-planar surfaces. In a real-world experiment, 1-D CNN achieved an overall accuracy of 87.5% in differentiating visually similar materials, demonstrating the effectiveness of transient histograms for material classification where conventional RGB-based methods fail. These findings highlight the potential of SPAD sensors combined with advanced classification techniques to enhance material classification and segmentation, providing a versatile framework for applications in robotics, computer vision, and optical sensing.https://ieeexplore.ieee.org/document/10942358/Material classificationsingle photon avalanche diodesurface segmentationtemporal resolutiontransient histogram1-D convolutional neural network
spellingShingle Yohanssen Pratama
Kazuya Kitano
Takahiro Kushida
Yuki Fujimura
Takuya Funatomi
Yasuhiro Mukaigawa
Material Segmentation Using 1-D Convolutional Neural Network With Transient Histogram
IEEE Access
Material classification
single photon avalanche diode
surface segmentation
temporal resolution
transient histogram
1-D convolutional neural network
title Material Segmentation Using 1-D Convolutional Neural Network With Transient Histogram
title_full Material Segmentation Using 1-D Convolutional Neural Network With Transient Histogram
title_fullStr Material Segmentation Using 1-D Convolutional Neural Network With Transient Histogram
title_full_unstemmed Material Segmentation Using 1-D Convolutional Neural Network With Transient Histogram
title_short Material Segmentation Using 1-D Convolutional Neural Network With Transient Histogram
title_sort material segmentation using 1 d convolutional neural network with transient histogram
topic Material classification
single photon avalanche diode
surface segmentation
temporal resolution
transient histogram
1-D convolutional neural network
url https://ieeexplore.ieee.org/document/10942358/
work_keys_str_mv AT yohanssenpratama materialsegmentationusing1dconvolutionalneuralnetworkwithtransienthistogram
AT kazuyakitano materialsegmentationusing1dconvolutionalneuralnetworkwithtransienthistogram
AT takahirokushida materialsegmentationusing1dconvolutionalneuralnetworkwithtransienthistogram
AT yukifujimura materialsegmentationusing1dconvolutionalneuralnetworkwithtransienthistogram
AT takuyafunatomi materialsegmentationusing1dconvolutionalneuralnetworkwithtransienthistogram
AT yasuhiromukaigawa materialsegmentationusing1dconvolutionalneuralnetworkwithtransienthistogram