Smart Segmentation Research Articles

Internet of things enabled diabetic foot ulcer image analysis support for smart segmentation using virtual sensing

Internet of things enabled diabetic foot ulcer image analysis support for smart segmentation using virtual sensing

Internet of things enabled diabetic foot ulcer image analysis support for smart segmentation using virtual sensing

Internet of things enabled diabetic foot ulcer image analysis support for smart segmentation using virtual sensing

Image Smart Segmentation Analysis Against Diabetic Foot Ulcer Using Internet of Things with Virtual Sensing.

Diabetic foot ulcer (DFU) is a problem worldwide, and prevention is crucial. The image segmentation analysis of DFU identification plays a significant role. This will produce different segmentation of the same idea, incomplete, imprecise, and other problems. To address these issues, a method of image segmentation analysis of DFU through internet of things with the technique of virtual sensing for semantically similar objects, the analysis of four levels of range segmentation (region-based, edge-based, image-based, and computer-aided design-based range segmentation) for deeper segmentation of images is implemented. In this study, the multimodal is compressed with the object co-segmentation for semantical segmentation. The result is predicting the better validity and reliability assessment. The experimental results demonstrate that the proposed model can efficiently perform segmentation analysis, with a lower error rate, than the existing methodologies. The findings on the multiple-image dataset show that DFU obtains an average segmentation score of 90.85% and 89.03% correspondingly in two types of labeled ratios before DFU with virtual sensing and after DFU without virtual sensing (i.e., 25% and 30%), which is an increase of 10.91% and 12.22% over the previous best results. In live DFU studies, our proposed system improved by 59.1% compared with existing deep segmentation-based techniques and its average image smart segmentation improvements over its contemporaries are 15.06%, 23.94%, and 45.41%, respectively. Proposed range-based segmentation achieves interobserver reliability by 73.9% on the positive test namely likelihood ratio test set with only a 0.25 million parameters at the pace of labeled data.

Abstract 5643: High-spatial-resolution multi-omics analysis of cancer tissues

Abstract Recent technological breakthroughs in single cell sequencing have revealed that malignant cells and nonmalignant cells are both highly dynamic and in the case of tumor cells, that the intratumoral heterogeneity is quite remarkable. However, several questions remain, especially in a spatial context, and a rapidly growing need has been observed for imaging-based approaches with single-cell resolution to visualize and characterize the tumor microenvironment. In the current study, we used Miltenyi Biotec newly developed spatial gene expression technology and automated imaging platform, to acquire spatial microscopy data for multiple biological analytes and multiple cancer tissues. The approach is based on fluorescence microscopy and the MICS technology principle of cyclic staining (MACSima™ Imaging Cyclic Staining) along with validated fluorochrome-conjugated antibodies and specifically amplified oligonucleotide probes combined with a proprietary signal removal system. FFPE tissue blocks were analyzed and each gene, which display a unique emission spectrum during a single cycle, was decoded. Due to the non-destructive nature of the workflow, H&E staining, immunofluorescence and RNA data can be collected together into a single image and on a single section. Using our antibody and targeted RNA immune-oncology panel coupled with our smart segmentation and flexible gating software suite, we have generated a topographical map of the expression profile level for several cancer tissues i.e.: tonsil, pancreas, kidney and ovary. This technology allows for the analysis of hundreds of protein markers and RNA transcripts on a single tissue, the characterization of the spatial architecture and the expression topography of tumor tissue sections. A better understanding of tissue microenvironment and implementation of spatial biology approaches will help advance the identification of more suitable biomarkers and for cancer therapy. Citation Format: Emily Neil, Dongju Park, Erica Lloyd, Michael Dibuono, Seiyu Hosono, Reto Muller, Hanna Lafayette, Hsinyi Smith, Robert Pinard. High-spatial-resolution multi-omics analysis of cancer tissues. [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2023; Part 1 (Regular and Invited Abstracts); 2023 Apr 14-19; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2023;83(7_Suppl):Abstract nr 5643.

Evaluation of EfficientNet models for COVID-19 detection using lung parenchyma.

When the COVID-19 pandemic broke out in the beginning of 2020, it became crucial to enhance early diagnosis with efficient means to reduce dangers and future spread of the viruses as soon as possible. Finding effective treatments and lowering mortality rates is now more important than ever. Scanning with a computer tomography (CT) scanner is a helpful method for detecting COVID-19 in this regard. The present paper, as such, is an attempt to contribute to this process by generating an open-source, CT-based image dataset. This dataset contains the CT scans of lung parenchyma regions of 180 COVID-19-positive and 86 COVID-19-negative patients taken at the Bursa Yuksek Ihtisas Training and Research Hospital. The experimental studies show that the modified EfficientNet-ap-nish method uses this dataset effectively for diagnostic purposes. Firstly, a smart segmentation mechanism based on the k-means algorithm is applied to this dataset as a preprocessing stage. Then, performance pretrained models are analyzed using different CNN architectures and with our Nish activation function. The statistical rates are obtained by the various EfficientNet models and the highest detection score is obtained with the EfficientNet-B4-ap-nish version, which provides a 97.93% accuracy rate and a 97.33% F1-score. The implications of the proposed method are immense both for present-day applications and future developments.

The Correlation of Body Mass Index with Visceral and Subcutaneous Adipose Tissue Volume in Patients that Performed Abdominal CT scan at RSUP H. Adam Malik Medan

Visceral and subcutaneous adipose tissue (VAT and SAT), and excess body mass index (BMI) act as cardiometabolic risk factors. BMI only describes the ratio of body weight to height in general. Abdominal CT scan is the gold standard in the examination of VAT and SAT volumes. The purpose of this study is to analyze the correlation between BMI with VAT and SAT in patients that performed CT scans of the abdomen. A total of 60 patients who underwent CT scans of the abdominal region at the Radiology Department of the RSUP H. Adam Malik Medan from July to August 2022 participated in this cross-sectional study. The BMI was calculated from measurement and volume of VAT and SAT were obtained from CT scans of the abdominal region at the fourth lumbal level, using 3D smart segmentation. Among of samples was 51.7% male. Average of age and BMI are 49.3 12.18 years and 25.58 kg/m2. Average of VAT and SAT volume value are 678.13 436.09 cm3 and 1708.76 1068.99 cm3. There is a moderate positive correlation between BMI and VAT (r=0.687) and BMI and SAT (r=0.669) with p<0.001.

Deep-layers-assisted machine learning for accurate image segmentation of complex materials

In materials science, X-ray micro-computed tomography (μCT) imagery provides a unique tool for quantitative and qualitative characterization of structure and performance. The tomographic analysis relies on the segmentation of large data sets to produce quantifiable volumes, a highly time-intensive process. Great progress has been observed in developing robust and efficient segmentation techniques, and machine learning currently stands out as a powerful segmentation tool. These methods have demonstrated excellent performance when applied to various materials. However, the performance of such methods heavily relies on the amount of ground truth data. This presents a great challenge in image segmentation since the amount of a priori raw images can be limited by sample cost and size. Additionally, manual labeling of data sets is often very time-intensive and can be subjected to human error. Therefore, a machine learning method is needed which achieves high segmentation accuracies for reduced training data sets. To this end, we have offered a stochastic method for producing variations of the training images that will retain the important class-wide features and thereby enrich the machine learning's “understanding” of the variabilities. The proposed method can significantly increase the final overall accuracy. We found that by enlarging the initial training set by additional realizations, we were able to improve the average accuracy of segmentation from 81.1% to 90.0% for a very complex Mg-based alloy. The results of this study show that it is possible to increase the accuracy of predictions in imaging from X-ray microscopy using machine learning methods when enough data are not available. • Rapid and smart segmentation of multiphase complex materials is studied. • We presented a machine learning method that is combined with a stochastic method. • Our method shows a significant improvement over the original machine learning methods. • The proposed method can be used when enough data is not available for machine learning.

A mathematical and dosimetric approach to validate auto‐contouring by Varian Smart segmentation for prostate cancer patients

AbstractPurposeThe aim of this study was to quantify the discrepancies in geometrical and dosimetric impacts (in volumetric modulated arc therapy) between manually segmented (MS) contours and smart segmentation (SS) auto‐contours (by Varian Eclipse Treatment Planning System SS v13.5) for prostate cancer patients.MethodsThe automated segmentation was carried out by Eclipse Treatment Planning System (Varian, version 13.5) Smart Segmentation (SS) workspace of 10 prostate cancer patients for four regions of interest; such as, bladder, rectum, femoral head left, and femoral head right. The geometric and dosimetric deviation between SS and MS contours have been quantified in the form of different parameters. The organ‐wise correlation between different validation parameters was addressed.ResultsThe organ‐wise correlation analysis showed the good and consistent correlation between different geometric validation parameters for the bladder. The hypothesis test for checking compliance of different parameters with AAPM 132 tolerance was addressed and validated between MS and SS bladder with p‐value = 0.01 and 0.05. There was no significant dosimetric difference between the dose–volume histogram (DVH) estimated for the SS bladder and standard DVH constraints protocol (as per the TMH PRIME trial) with p‐value = 0.01 and 0.05. The difference between DVH estimated for MS and SS bladder was also not significant, with p‐value = 0.05.ConclusionThis study shows that “well correlated validation parameters infer correctly about the matching or coincidence between auto and manually segmented contours,” and the bladder contouring by Smart Segmentation and plan optimization can achieve acceptable DVH constraints.

IoT 네트워크에서 보안위협 확산 방지를 위한 스마트 세그멘테이션 프레임워크 구현

이동통신과 사물인터넷 발전에 따라 네트워크 연결 디바이스 증가 및 사물인터넷 서비스 대중화로 디바이스 취약점을 악용하는 보안사고가 급증하고 있다. 특히 제한된 자원 특성으로 보안이 취약한 디바이스들이 해커들의 공격 대상이 되고 있으며, 대표적인 것이 대규모 분산 서비스 거부(DDoS) 공격을 유발하는 Mirai 악성코드의 감염확산이다. DDoS 공격 대응 방법 중 하나는 DDoS 공격 원인인 악성코드 확산 억제를 통한 대규모 사물봇넷 생성을 방지하는 것이다. 본 논문은 IoT 네트워크 내의 보안위협 확산을 억제하기 위해 디바이스 속성, 네트워크 정보, 서비스 유형 등을 기준으로 세그멘트를 정의하고, 보안위협 확산 여부에 따라 세그멘트 단위로 네트워크 접속을 자율적으로 통제하는 스마트 세그멘테이션 프레임워크에 대한 구현을 설명한다. 제안된 스마트 세그멘테이션 프레임워크는 Mirai 악성코드 감염 환경에서의 시험을 통해 악성코드 확산을 억제할 수 있음을 검증하였다.

Smart Approach for Glioma Segmentation in Magnetic Resonance Imaging using Modified Convolutional Network Architecture (U-NET)

Segmentation of a brain tumor from magnetic resonance multimodal images is a challenging task in the field of medical imaging. The vast diversity in potential target regions, appearance and multifarious intensity threshold levels of various tumor types are few of the major factors that affect segmentation results. An accurate diagnosis and its treatment demand strict delineation of the tumor affected tissues. Herein, we focus on a smart, automated, and robust segmentation approach for brain tumor using a modified 3D U-Net architecture. The pre-operative multimodal 3D-MRI scans of High-Grade Glioma (HGG) and Low-Grade Glioma (LGG) are used as data. Our proposed approach solves the problem of memory and system resource constraints by robustly applying dense network training on image patches of 3D volumes. It improves the border region artifact detection by applying convolutions at an appropriate phase in the proposed neural network. Multi-class imbalance data are handled by using Categorical Cross Entropy (CCE) loss developed by combining the Weighted Cross Entropy (WCE) with Weighted Multi-class Dice Loss (WMDL) functions, which enables the network to perform smart segmentation of the smaller tumorous regions. The proposed approach is tested and evaluated for the challenge datasets of multimodal MRI volumes of tumor patients. Experiments are performed to compute the average dice scores on BraTS-2019 and BraTS-2020 datasets for the whole tumor region.

Efficiency of «smart segments» in peripheral and face diamond grinding of abrasive and non-abrasive brittle materials

A comparative analysis of the machining of the peripheral and end parts of a diamond tool is presented. The relevance of using modular segments for face grinding is shown. The features of their design in the tool for face grinding of products from solid materials: concrete, ceramics and natural stone are considered. A conditional classification of materials during processing at face grinding operations for abrasive and non-abrasive is proposed. On the basis of modular «smart segments», a gamut of high-performance and resistant tools has been created for sequential processing from the calibration operation to fine grinding of the surface of both classes of materials. Special designs of replaceable (interchangeable) grinding elements for equipping stationary and mobile grinding installations are proposed.

Effectiveness of “Smart Segments” in Peripheral and End Diamond Grinding of Abrasive and Non-Abrasive Brittle Materials

Effectiveness of “Smart Segments” in Peripheral and End Diamond Grinding of Abrasive and Non-Abrasive Brittle Materials

Automatic segmentation of organs at risk for nasopharyngeal carcinoma with Smart Segmentation and MIM Atlas

Objective To compare the accuracy of two automatic segmentation softwares (Smart Segmentation and MIM Atlas) in organs at risk (OARs) contouring for nasopharyngeal carcinoma (NPC). Methods Totally 55 NPC patients were retrospectively reviewed with manually contoured OARs on CT images, in which 30 cases were randomly selected to create a data base in the Smart Segmentation and MIM Atlas. The remaining 25 cases were automatically contoured with Smart Segmentation and MIM as test cases. The automatic contouring accuracies of two softwares were evaluated with Dice coefficient(DSC), Hausdorff distance(HD), and absolute volume difference(△V) using manual contours as a golden standard. Results The overall DSC, HD and △V of all organs contoured by MIM Atlas and Smart Segmentation were (0.79±0.13) vs. (0.62±0.24) (t=14.06, P<0.05), (5.50±3.84)mm vs.(8.38±4.88)mm (t=-11.40, P<0.05), and (1.52±2.46) cm3vs. (2.38±3.57) cm3 (t=-4.70, P<0.05), respectively. The average DSC of 11 organs (brain stem, optic chiasm, bilateral lens, bilateral optic nerve, bilateral eyeballs, bilateral parotid gland, spinal cord) delineated by MIM Atlas was statistically greater than that of Smart Segmentation (t=5.27, 4.41, 6.34, 5.70, 10.62, 7.45, 3.96, 4.26, 6.25, 5.42, 7.23, P<0.05). The average HD of 10 organs (brain stem, optic chiasm, bilateral lens, bilateral optic nerve, bilateral eyeballs, left parotid gland, spinal cord) delineated by MIM Atlas was statistically less than that of Smart Segmentation (t=-4.51, -4.49, -3.92, -3.45, -5.36, -5.56, -3.89, -3.90, -3.60, -3.68, P<0.05). The average △V of 6 organs (brain stem, optic chiasm, left len, bilateral optic nerve, right eyeball) delineated by MIM Atlas was statistically less than that of Smart Segmentation (t=-2.83, -3.39, -2.56, -2.27, -2.43, -2.51, P<0.05). Conclusions Both softwares have reasonable contouring accuracy for larger organs. The accuracy decreased with the decrease of organ volumes and blurred boundary. Generally, MIM Atlas′s performs better than Smart Segmentation does. Key words: Automatic contouring; Organs-at-risk segmentation; Atlas library; Nasopharyngeal carcinoma

The Histochemical Alterations of Mucin in Colorectal Carcinoma Quantified by Two Efficient Algorithms of Digital Image Analysis.

The practical use of knowledge on the diagnostic-prognostic role of polysaccharide components of mucins in colorectal cancer (CRC) has been difficult, due to the number of histochemical (HC) reaction types, as well as lack of standard methods of computer-assisted analysis of tissue expression of these molecules. Using two algorithms of digital image analysis (by application of Image-Pro Premier and our originally designed program Filter HSV), we evaluated the expression of polysaccharides in tissue samples of CRC patients (n = 33), and fragments of normal colorectal tissue from the same patients (control) using periodic acid Schiff reaction (PAS) (neutral mucins) and alcian blue staining (AB) (acidic mucins). Our results indicate lower expression of the PAS+ and AB+ mucins in CRC, as compared to the control samples. The higher expression of PAS+ polysaccharides was detected in flat tumors than in protruded CRC, while higher AB+ mucins expression was a feature of mucinous CRC subtypes. Positive correlation between mutual PAS+ and AB+ expression, as well as correlations with glucose concentration (PAS+ mucins), and hemoglobin level (AB+ mucins) were observed exclusively in unchanged colorectal samples (control). Both algorithms of digital image analysis (smart segmentation and Filter HSV) work properly and can be used interchangeably in daily practice of pathologists, as useful tools of quantitative evaluation of HC reaction in both normal and cancerous tissues.

Automatic Intracranial Segmentation: Is the Clinician Still Needed?

Introduction:Stereotactic hypofractionated radiotherapy is an effective treatment for brain metastases in oligometastatic patients. Its planning is however time-consuming because of the number of organs at risk to be manually segmented. This study evaluates 2 automated segmentation commercial software.Methods:Patients were scanned in the treatment position. The computed tomography scan was registered on a magnetic resonance imaging and volumes were manually segmented by a clinician. Then 2 automated segmentations were performed (with iPlan and Smart Segmentation). RT STRUCT files were compared with Aquilab’s Artistruct segment comparison module. We selected common segmented volume ratio as the main judging criterion. Secondary criteria were Dice-Sørensen coefficients, overlap ratio, and additional segmented volume.Results:Twenty consecutive patients were included. Agreement between manual and automated contouring was poor. Common segmented volumes ranged from 7.71% to 82.54%, Dice-Sørensen coefficient ranged from 0.0745 to 0.8398, overlap ratio ranged from 0.0414 to 0.7275, and additional segmented volume ranged from 9.80% to 92.25%. Each software outperformed the other on some organs while performing worse on others.Conclusion:No software seemed clearly better than the other. Common segmented volumes were much too low for routine use in stereotactic hypofractionated brain radiotherapy. Manual editing is still needed.

Segments of A Diamond Tool with Compound Structure for Cutting of Hard Brittle Materials – an Approach to the Creation of “Smart Segments”1

Certain principles for the construction of the diamond component of segments of a cutting and grinding tool for use in cutting of hard brittle materials, including refractories, are considered. Examples of constructions of segments designed for a particular function are presented. Principles for the construction of compound “smart segments” that exert additional effects in the cutting zone (besides cutting itself) which serve to adapt the cutting conditions to the tool are presented. Examples of cutting of different materials that demonstrate the universality and efficiency of the proposed segments are presented.

Abstract 425: Nuclear morphometry predicts prostate cancer progression

Abstract Introduction: The progression of prostate cancer (PCa) involves tissue morphometric changes, which laid the base for current pathological diagnosis of prostate cancer with the Gleason Scoring system. Nuclear morphometry (NM) changes contribute significantly to these tissue morphometric changes, but accurate and autonomous quantification of NM changes are challenging. Here we created a macro to quantify the changes of size, shape, DNA content, etc., and used these parameters to predict PCa progression in 80 men that underwent radical prostatectomy (RP). Materials &amp; Methods: Two tissue microarrays (TMAs) with 80 RP PCa cases, stratified by Gleason scores (GS) were used for this study. The two continuous sections of TMAs were stained with H&amp;E and Feulgen reagents and the H&amp;E slides were used for pathological diagnosis of PCa. Both kinds of slides were scanned with Aperio scanner and image of each core were separated using Aperio ImageScope software. The H&amp;E or Feulgen stained nuclei of cancer core were quantified using Smart Segmentation of ImagePro® Premier 9.1 software. For each core, data of all ROIs were pooled and the covariance of each parameter were generated. Data were first analyzed alone and then in combination using multivariate logistic regression (MLR) to predict the aggressive RP cases or biochemical recurrence (BCR). Decision curve analysis (DCA) was used for the MLR models to evaluate their power and effectiveness for clinical decision making. For all analysis, a p&amp;lt;0.05 is considered as statistically significant. Results: Using multivariate logistic regression (MLR), to differentiate aggressive PCa (Gleason score 4+3 &amp; &amp;gt; = 8) from less aggressive PCa (Gleason score 3+3 &amp; 3+4) on the TMAs of RP cases, our Feulgen NM model generated a receiver operating characteristic curve-area under the curve (ROC-AUC) of 0.90 with a sensitivity of 75.51% and specificity of 86.21% while our H&amp;E NM model yield an ROC-AUC of 0.96 with a sensitivity of 82.00% and specificity of 90.00%. DCA analysis both showed better decision using both models compared with considering the patients as all or none of them as aggressive. Both models also show strength in differentiating two Gleason score 7 (3+4 vs 4+3). Feulgen NM model showed moderate power in the prediction of BCR based on the MLR (ROC-AUC = 0.79) and DCA analysis, while H&amp;E MLR model performed better with a ROC-AUC of 0.86. Further Kaplan-Meier analysis of BCR with GS and NM (H&amp;E and Feulgen) showed that the combination of GS with either Feulgen or H&amp;E NM showed significant power in the differentiation patients with BCR survival time (HR = 3.14 &amp; 3.89, respectively). Conclusions: Our accurate quantification of tissue morphometry demonstrated its translational clinical relevance since it can predict PCa aggressiveness in men that have undergone RP. Future application of this tool in active surveillance biopsies may provide an early diagnosis and prognosis prediction of PCa patients effectively. Citation Format: Guangjing Zhu, Aniq ur rehman Gajdhar, Jonathan I. Epstein, Neil Carleton, Christine Davis, Luciane Tsukamoto Kagohara, Robert W. Veltri. Nuclear morphometry predicts prostate cancer progression. [abstract]. In: Proceedings of the 107th Annual Meeting of the American Association for Cancer Research; 2016 Apr 16-20; New Orleans, LA. Philadelphia (PA): AACR; Cancer Res 2016;76(14 Suppl):Abstract nr 425.

A smart segmentation technique towards improved infrequent non-speech gestural activity recognition model

Infrequent Non-Speech Gestural Activities (IGAs) such as coughing, deglutition and yawning help identify fine-grained physiological symptoms and chronic psychological conditions which are not directly observable from traditional daily activities. We propose a new wearable smart earring which is capable of differentiating IGAs in daily environment with single integrated accelerometer sensor signal processing. Our prior framework, GeSmart, shows significant improvement in IGAs recognition based on smart earring which necessitates users to replace the earring batteries frequently due to its energy hungry requirement (high sampling frequency) towards fine-grained IGAs recognition. In this improved work, we propose a new segmentation technique along with GeSmart which takes the advantages of change-point detection algorithm to segment sensor data streams, feature extraction and classification thus any machine learning technique can perform significantly well in low sampling rate. We also implement a baseline traditional graphical model based gesture recognition techniques and compare their performances with our model in terms of accuracy, energy consumption and degradation of sampling rate scenarios. Experimental results based on real data traces demonstrate that our approach improves the performances significantly compared to previously proposed solutions. We also apply our segmentation technique on two benchmark datasets to prove the superiority of our technique in low sampling rate scenario.

TU‐H‐CAMPUS‐JeP1‐02: Fully Automatic Verification of Automatically Contoured Normal Tissues in the Head and Neck

Purpose:To investigate and validate the use of an independent deformable‐based contouring algorithm for automatic verification of auto‐contoured structures in the head and neck towards fully automated treatment planning.Methods:Two independent automatic contouring algorithms [(1) Eclipse's Smart Segmentation followed by pixel‐wise majority voting, (2) an in‐house multi‐atlas based method] were used to create contours of 6 normal structures of 10 head‐and‐neck patients. After rating by a radiation oncologist, the higher performing algorithm was selected as the primary contouring method, the other used for automatic verification of the primary. To determine the ability of the verification algorithm to detect incorrect contours, contours from the primary method were shifted from 0.5 to 2cm. Using a logit model the structure‐specific minimum detectable shift was identified. The models were then applied to a set of twenty different patients and the sensitivity and specificity of the models verified.Results:Per physician rating, the multi‐atlas method (4.8/5 point scale, with 3 rated as generally acceptable for planning purposes) was selected as primary and the Eclipse‐based method (3.5/5) for verification. Mean distance to agreement and true positive rate were selected as covariates in an optimized logit model. These models, when applied to a group of twenty different patients, indicated that shifts could be detected at 0.5cm (brain), 0.75cm (mandible, cord), 1cm (brainstem, cochlea), or 1.25cm (parotid), with sensitivity and specificity greater than 0.95. If sensitivity and specificity constraints are reduced to 0.9, detectable shifts of mandible and brainstem were reduced by 0.25cm. These shifts represent additional safety margins which might be considered if auto‐contours are used for automatic treatment planning without physician review.Conclusion:Automatically contoured structures can be automatically verified. This fully automated process could be used to flag auto‐contours for special review or used with safety margins in a fully automatic treatment planning system.

A variational based smart segmentation model for speckled images

In this paper, we propose a new variational model in the fuzzy framework to achieve the task of partitioning speckled images, such as the synthetic aperture radar (SAR) images. The model is partly derived by using the so-called maximizing a posteriori (MAP) estimation method. The novelties of the model are that (1) the Gamma distribution rather than the classical Gaussian distribution is used to simulate the gray intensities in each homogeneous region of the images; (2) a smart regularization term with respect to fuzzy membership functions is designed. The newly designed regularization term equals to an adaptive weighted total variation (TV) regularizer. Compared with the classical TV regularizer, the proposed regularization term not only has a sparser property, but also protects the segmentation results from degeneration (being over-smoothed). In addition, a new algorithm based on the alternative direction iteration algorithm is proposed to solve the model. The algorithm is efficient since it integrates the split Bregman method and Chambolle׳s projection method. Numerical examples are given to verify the promising efficiency of our model.