The Experts below are selected from a list of 81609 Experts worldwide ranked by ideXlab platform

Matti Pietikainen - One of the best experts on this subject based on the ideXlab platform.

  • unsupervised texture Segmentation using feature distributions
    1999
    Co-Authors: Timo Ojala, Matti Pietikainen
    Abstract:

    Abstract This paper presents an unsupervised texture Segmentation method, which uses distributions of local binary patterns and pattern contrasts for measuring the similarity of adjacent image regions during the Segmentation Process. Nonparametric log-likelihood test, the G statistic, is engaged as a pseudo-metric for comparing feature distributions. A region-based algorithm is developed for coarse image Segmentation and a pixelwise classification scheme for improving localization of region boundaries. The performance of the method is evaluated with various types of test images.

  • unsupervised texture Segmentation using feature distributions
    1997
    Co-Authors: Timo Ojala, Matti Pietikainen
    Abstract:

    This paper presents an unsupervised texture Segmentation method, which uses distributions of local binary patterns and pattern contrasts for measuring the similarity of adjacent image regions during the Segmentation Process. Nonparametric log-likelihood test, the G statistic, is engaged as a pseudo-metric for comparing feature distributions. A region-based algorithm is developed for coarse image Segmentation and a pixelwise classification scheme for improving localization of region boundaries. The performance of the method is evaluated with various types of test images. The same set of parameter values is used in all the experiments with texture mosaics in order to demonstrate the robustness of our approach.

Peter O. Gerrits - One of the best experts on this subject based on the ideXlab platform.

  • The influence of the Segmentation Process on 3D measurements from cone beam computed tomography-derived surface models.
    2013
    Co-Authors: Willem P Engelbrecht, Zacharias Fourie, Janalt Damstra, Peter O. Gerrits, Yijin Ren
    Abstract:

    To compare the accuracy of linear and angular measurements between cephalometric and anatomic landmarks on surface models derived from 3D cone beam computed tomography (CBCT) with two different Segmentation protocols was the aim of this study. CBCT scans were made of cadaver heads and 3D surface models were created of the mandible using two different Segmentation protocols. A high-resolution laser surface scanner was used to make a 3D model of the macerated mandibles. Twenty linear measurements at 15 anatomic and cephalometric landmarks between the laser surface scan and the 3D models generated from the two Segmentation protocols (commercial Segmentation (CS) and doctor's Segmentation (DS) groups) were measured. The interobserver agreement for all the measurements of the all three techniques was excellent (intraclass correlation coefficient 0.97-1.00). The results are for both groups very accurate, but only for the measurements on the condyle and lingual part of the mandible, the measurements in the CS group is slightly more accurate than the DS group. 3D surface models produced by CBCT are very accurate but slightly inferior to reality when threshold-based methods are used. Differences in the Segmentation Process resulted in significant clinical differences between the measurements. Care has to be taken when drawing conclusions from measurements and comparisons made from different Segmentations, especially at the condylar region and the lingual side of the mandible.

  • Segmentation Process significantly influences the accuracy of 3d surface models derived from cone beam computed tomography
    2012
    Co-Authors: Zacharias Fourie, Janalt Damstra, R H Schepers, Peter O. Gerrits
    Abstract:

    AIMS: To assess the accuracy of surface models derived from 3D cone beam computed tomography (CBCT) with two different Segmentation protocols. MATERIALS AND METHODS: Seven fresh-frozen cadaver heads were used. There was no conflict of interests in this study. CBCT scans were made of the heads and 3D surface models were created of the mandible using two different Segmentation protocols. The one series of 3D models was segmented by a commercial software company, while the other series was done by an experienced 3D clinician. The heads were then macerated following a standard Process. A high resolution laser surface scanner was used to make a 3D model of the macerated mandibles, which acted as the reference 3D model or "gold standard". The 3D models generated from the two rendering protocols were compared with the "gold standard" using a point-based rigid registration algorithm to superimpose the three 3D models. The linear difference at 25 anatomic and cephalometric landmarks between the laser surface scan and the 3D models generate from the two rendering protocols was measured repeatedly in two sessions with one week interval. RESULTS: The agreement between the repeated measurement was excellent (ICC=0.923-1.000). The mean deviation from the gold standard by the 3D models generated from the CS group was 0.330mm±0.427, while the mean deviation from the Clinician's rendering was 0.763mm±0.392. The surface models segmented by both CS and DS protocols tend to be larger than those of the reference models. In the DS group, the biggest mean differences with the LSS models were found at the points ConLatR (CI: 0.83-1.23), ConMedR (CI: -3.16 to 2.25), CoLatL (CI: -0.68 to 2.23), Spine (CI: 1.19-2.28), ConAntL (CI: 0.84-1.69), ConSupR (CI: -1.12 to 1.47) and RetMolR (CI: 0.84-1.80). CONCLUSION: The Commercially segmented models resembled the reality more closely than the Doctor's segmented models. If 3D models are needed for surgical drilling guides or surgical planning which requires high precision, the additional cost of the commercial Segmentation services seem to be justified to produce a more accurate surface models.

Timo Ojala - One of the best experts on this subject based on the ideXlab platform.

  • unsupervised texture Segmentation using feature distributions
    1999
    Co-Authors: Timo Ojala, Matti Pietikainen
    Abstract:

    Abstract This paper presents an unsupervised texture Segmentation method, which uses distributions of local binary patterns and pattern contrasts for measuring the similarity of adjacent image regions during the Segmentation Process. Nonparametric log-likelihood test, the G statistic, is engaged as a pseudo-metric for comparing feature distributions. A region-based algorithm is developed for coarse image Segmentation and a pixelwise classification scheme for improving localization of region boundaries. The performance of the method is evaluated with various types of test images.

  • unsupervised texture Segmentation using feature distributions
    1997
    Co-Authors: Timo Ojala, Matti Pietikainen
    Abstract:

    This paper presents an unsupervised texture Segmentation method, which uses distributions of local binary patterns and pattern contrasts for measuring the similarity of adjacent image regions during the Segmentation Process. Nonparametric log-likelihood test, the G statistic, is engaged as a pseudo-metric for comparing feature distributions. A region-based algorithm is developed for coarse image Segmentation and a pixelwise classification scheme for improving localization of region boundaries. The performance of the method is evaluated with various types of test images. The same set of parameter values is used in all the experiments with texture mosaics in order to demonstrate the robustness of our approach.

Zacharias Fourie - One of the best experts on this subject based on the ideXlab platform.

  • The influence of the Segmentation Process on 3D measurements from cone beam computed tomography-derived surface models.
    2013
    Co-Authors: Willem P Engelbrecht, Zacharias Fourie, Janalt Damstra, Peter O. Gerrits, Yijin Ren
    Abstract:

    To compare the accuracy of linear and angular measurements between cephalometric and anatomic landmarks on surface models derived from 3D cone beam computed tomography (CBCT) with two different Segmentation protocols was the aim of this study. CBCT scans were made of cadaver heads and 3D surface models were created of the mandible using two different Segmentation protocols. A high-resolution laser surface scanner was used to make a 3D model of the macerated mandibles. Twenty linear measurements at 15 anatomic and cephalometric landmarks between the laser surface scan and the 3D models generated from the two Segmentation protocols (commercial Segmentation (CS) and doctor's Segmentation (DS) groups) were measured. The interobserver agreement for all the measurements of the all three techniques was excellent (intraclass correlation coefficient 0.97-1.00). The results are for both groups very accurate, but only for the measurements on the condyle and lingual part of the mandible, the measurements in the CS group is slightly more accurate than the DS group. 3D surface models produced by CBCT are very accurate but slightly inferior to reality when threshold-based methods are used. Differences in the Segmentation Process resulted in significant clinical differences between the measurements. Care has to be taken when drawing conclusions from measurements and comparisons made from different Segmentations, especially at the condylar region and the lingual side of the mandible.

  • Segmentation Process significantly influences the accuracy of 3d surface models derived from cone beam computed tomography
    2012
    Co-Authors: Zacharias Fourie, Janalt Damstra, R H Schepers, Peter O. Gerrits
    Abstract:

    AIMS: To assess the accuracy of surface models derived from 3D cone beam computed tomography (CBCT) with two different Segmentation protocols. MATERIALS AND METHODS: Seven fresh-frozen cadaver heads were used. There was no conflict of interests in this study. CBCT scans were made of the heads and 3D surface models were created of the mandible using two different Segmentation protocols. The one series of 3D models was segmented by a commercial software company, while the other series was done by an experienced 3D clinician. The heads were then macerated following a standard Process. A high resolution laser surface scanner was used to make a 3D model of the macerated mandibles, which acted as the reference 3D model or "gold standard". The 3D models generated from the two rendering protocols were compared with the "gold standard" using a point-based rigid registration algorithm to superimpose the three 3D models. The linear difference at 25 anatomic and cephalometric landmarks between the laser surface scan and the 3D models generate from the two rendering protocols was measured repeatedly in two sessions with one week interval. RESULTS: The agreement between the repeated measurement was excellent (ICC=0.923-1.000). The mean deviation from the gold standard by the 3D models generated from the CS group was 0.330mm±0.427, while the mean deviation from the Clinician's rendering was 0.763mm±0.392. The surface models segmented by both CS and DS protocols tend to be larger than those of the reference models. In the DS group, the biggest mean differences with the LSS models were found at the points ConLatR (CI: 0.83-1.23), ConMedR (CI: -3.16 to 2.25), CoLatL (CI: -0.68 to 2.23), Spine (CI: 1.19-2.28), ConAntL (CI: 0.84-1.69), ConSupR (CI: -1.12 to 1.47) and RetMolR (CI: 0.84-1.80). CONCLUSION: The Commercially segmented models resembled the reality more closely than the Doctor's segmented models. If 3D models are needed for surgical drilling guides or surgical planning which requires high precision, the additional cost of the commercial Segmentation services seem to be justified to produce a more accurate surface models.

Kai Liu - One of the best experts on this subject based on the ideXlab platform.

  • thyroid nodule Segmentation in ultrasound images based on cascaded convolutional neural network
    2018
    Co-Authors: Xiang Ying, Mankun Zhao, Kai Liu
    Abstract:

    Based on U-shaped Fully Convolutional Neural Network (UNET), Convolutional Neural Network (CNN) classifier and Deep Fully Convolutional Neural Network (FCN), this paper proposes a thyroid nodule Segmentation model in form of cascaded convolutional neural network. In this paper, we study the Segmentation of thyroid nodules from two aspects, Segmentation Process and model structure. On the one hand, the research of the Segmentation Process includes the gradual reduction of the Segmentation region and the selection of different model structures. On the other hand, the research of model structures includes the design of network structure, the adjustment of model parameters and so on. And the experiment shows that our thyroid nodule Segmentation in ultrasound images has a good performance, which is superior to the current algorithms and can be used as a reference for the diagnosis of the doctor.