Multiple Segmentation Algorithms Research Articles

A shape-based multiple segmentation algorithm for change-point detection

We consider the detection and localization of change points for the off-line sequence of observations. Specifically, we propose a new multi-segmentation algorithm for detecting multiple change-points, named shape-based multiple segmentation algorithm, which is a generalization of binary segmentation. The proposed method is combined with deep mining on the shape information of the test statistics curve to overcome the Gaussian distribution hypothesis limitation and the limitation of traditional segmentation methods only being able to detect one change-point per stage. Combined with shape context, a robust testing statistic was developed via a shape-based descriptor statistic instead of the traditional CUSUM statistic. Then a data-driven threshold by the rightmost sudden-drop point is proposed, and the change points are further identified by single-peak identification. An efficient multiple segmentation based on a shape recognition procedure is implemented to locate change points. The effectiveness of the proposed procedure is illustrated using both synthetic data sets and real world data from electrical distribution networks.

Deep learning to detect bacterial colonies for the production of vaccines

During the development of vaccines, bacterial colony forming units (CFUs) are counted in order to quantify the yield in the fermentation process. This manual task is long, tedious, and subject to errors. In this work, multiple segmentation algorithms based on the U-Net CNN architecture are tested and proven to offer robust, automated CFU counting. It is also shown that the multiclass generalisation with a bespoke loss function allows virulent and avirulent colonies to be distinguished with acceptable accuracy. While many possibilities are left to explore, our results show the potential of deep learning for separating and classifying bacterial colonies.

Quality-based iris segmentation-level fusion

Iris localisation and segmentation are challenging and critical tasks in iris biometric recognition. Especially in non-cooperative and less ideal environments, their impact on overall system performance has been identified as a major issue. In order to avoid a propagation of system errors along the processing chain, this paper investigates iris fusion at segmentation-level prior to feature extraction and presents a framework for this task. A novel intelligent reference method for iris segmentation-level fusion is presented, which uses a learning-based approach predicting ground truth segmentation performance from quality indicators and model-based fusion to create combined boundaries. The new technique is analysed with regard to its capability to combine segmentation results (pupillary and limbic boundaries) of multiple segmentation algorithms. Results are validated on pairwise combinations of four open source iris segmentation algorithms with regard to the public CASIA and IITD iris databases illustrating the high versatility of the proposed method.

An effective hand vein feature extraction method

As a new authentication method developed years ago, vein recognition technology features the unique advantage of bioassay. This paper studies the specific procedure for the extraction of hand back vein characteristics. There are different positions used in the collecting process, so that a suitable intravenous regional orientation method is put forward, allowing the positioning area to be the same for all hand positions. In addition, to eliminate the pseudo vein area, the valley regional shape extraction operator can be improved and combined with multiple segmentation algorithms. The images should be segmented step by step, making the vein texture to appear clear and accurate. Lastly, the segmented images should be filtered, eroded, and refined. This process helps to filter the most of the pseudo vein information. Finally, a clear vein skeleton diagram is obtained, demonstrating the effectiveness of the algorithm. This paper presents a hand back vein region location method. This makes it possible to rotate and correct the image by working out the inclination degree of contour at the side of hand back.

Weakly-supervised object localization in unlabeled image collection

Fully annotated image dataset is required for supervised learning. However, the image labeling process is laborious and monotonous. In this paper, we focus on automatic image labeling for a class-specified image dataset. We propose a weakly supervised approach to localize objects in a class of unlabelled images without using any manually labeled examples. Firstly, an image is segmented based on a multiple segmentation algorithm. Secondly, the segmented regions are mined based on the commonality and saliency to discovery the category pattern in the image. Thirdly, objects are localized based on the weakly supervised learning algorithm. To prove the effectiveness of the proposed approach, we experimentally evaluate the performance of our approach on 12 object classes of the Caltech101 dataset and 2 landmark classes collected from the Internet. The experimental results demonstrate that our approach is effective and accurate to automatically label images.

Comparison of image segmentation methods in simulated 2D and 3D microtomographic images of soil aggregates

Advances in X-ray microtomography (μCT) are opening new opportunities for examining soil pore structures. However, usefulness of μCT data for pore structure characterization depends on how accurately the grayscale images are segmented into pore and solid components. Multiple segmentation algorithms have been developed; however, one of the difficulties in comparing the accuracy of segmentation algorithms is the lack of ground-truth information in the soil samples subjected to μCT. This means that only the criteria that do not depend on the availability of the ground-truth data can be used in assessing accuracy of the segmentation methods, yet the reliability of such criteria in soil images is unclear. In this study, we simulated 2D and 3D soil images to resolve the problem of the lack of ground-truth information. The objectives of the study were (i) to explore optimal parameter selection for indicator kriging (IK) segmentation; (ii) to compare the accuracy of several commonly used segmentation methods, namely, entropy based method, iterative method, Otsu's method, and IK method; and (iii) to evaluate performance of the region non-uniformity measure (NU), the criterion that does not depend on presence of the ground-truth image, in segmentation method selection for soil images. We found that though there was no single segmentation method that preserved pore characteristics in all the cases, IK method yielded segmented images most similar to the ground-truth in most of the cases when the histogram of image grayscale values had clearly distinguishable peaks. For the image with poorly distinguishable histogram peaks, IK did not perform well, while Otsu's method produced acceptable segmentation results. The results indicated that selecting the segmentation method based on NU did not always produce optimal representation of pore characteristics. However, overall, the NU was found to be an acceptable criterion for segmentation method selection in μCT soil images.

Unsupervised identification of useful visual landmarks using multiple segmentations and top-down feedback

In this paper, we tackle the problem of unsupervised selection and posterior recognition of visual landmarks in image sequences acquired by an indoor mobile robot. This is a highly valuable perceptual capability for a wide variety of robotic applications, in particular autonomous navigation. Our method combines a bottom-up data driven approach with top-down feedback provided by high level semantic representations. The bottom-up approach is based on three main mechanisms: visual attention, area segmentation, and landmark characterization. As there is no segmentation method that works properly in every situation, we integrate multiple segmentation algorithms in order to increase robustness of the approach. In terms of top-down feedback, this is provided by two information sources: (i) An estimation of the robot position that reduces the searching scope for potential matches with previously selected landmarks, (ii) A set of weights that, according to the results of previous recognitions, controls the influence of each segmentation algorithm in the recognition of each landmark. We test our approach with encouraging results in three datasets corresponding to real-world scenarios.

Model-based detection of lung nodules in computed tomography exams 1: Thoracic computer-aided diagnosis

Rationale and objectives In this study, we developed a prototype model-based computer aided detection (CAD) system designed to automatically detect both solid and subsolid pulmonary nodules in computed tomography (CT) images. By using this CAD algorithm, along with the radiologist’s initial interpretation, we aim to improve the sensitivity of radiologic readings of CT lung exams. Materials and methods We have developed a model-based CAD algorithm through the use of precise mathematic models that capture scanner physics and anatomic information. Our model-based CAD algorithm uses multiple segmentation algorithms to extract noteworthy structures in the lungs and a Bayesian statistical model selection framework to determine the probability of various anatomical events throughout the lung. We tested this algorithm on 50 low-dose CT lung cancer screening cases in which ground truth was produced through readings by three expert chest radiologists. Results Using this model-based CAD algorithm on 50 low-dose CT cases, we measured potential sensitivity improvements of 7% and 5% in two radiologists with respect to all noncalcified nodules, solid and subsolid, greater than 5 mm in diameter. The third radiologist did not miss any nodules in the ground truth set. The CAD algorithm produced 8.3 false positives per case. Conclusion Our prototype CAD system demonstrates promising results as a tool to improve the quality of radiologic readings by increasing radiologist sensitivity. A significant advantage of this model-based approach is that it can be easily extended to support additional anatomic models as clinical understanding and scanning practices improve.