Multi-scale Texture Features Research Articles

A novel breast cancer image classification model based on multiscale texture feature analysis and dynamic learning

Assistive medical image classifiers can greatly reduce the workload of medical personnel. However, traditional machine learning methods require large amounts of well-labeled data and long learning times to solve medical image classification problems, which can lead to high training costs and poor applicability. To address this problem, a novel unsupervised breast cancer image classification model based on multiscale texture analysis and a dynamic learning strategy for mammograms is proposed in this paper. First, a gray-level cooccurrence matrix and Tamura coarseness are used to transfer images to multiscale texture feature vectors. Then, an unsupervised dynamic learning mechanism is used to classify these vectors. In the simulation experiments with a resolution of 40 pixels, the accuracy, precision, F1-score and AUC of the proposed method reach 91.500%, 92.780%, 91.370%, and 91.500%, respectively. The experimental results show that the proposed method can provide an effective reference for breast cancer diagnosis.

Least square-support vector machine based brain tumor classification system with multi model texture features

Radiologists confront formidable challenges when confronted with the intricate task of classifying brain tumors through the analysis of MRI images. Our forthcoming manuscript introduces an innovative and highly effective methodology that capitalizes on the capabilities of Least Squares Support Vector Machines (LS-SVM) in tandem with the rich insights drawn from Multi-Scale Morphological Texture Features (MMTF) extracted from T1-weighted MR images. Our methodology underwent meticulous evaluation on a substantial dataset encompassing 139 cases, consisting of 119 cases of aberrant tumors and 20 cases of normal brain images. The outcomes we achieved are nothing short of extraordinary. Our LS-SVM-based approach vastly outperforms competing classifiers, demonstrating its dominance with an exceptional accuracy rate of 98.97%. This represents a substantial 3.97% improvement over alternative methods, accompanied by a notable 2.48% enhancement in Sensitivity and a substantial 10% increase in Specificity. These results conclusively surpass the performance of traditional classifiers such as Support Vector Machines (SVM), Radial Basis Function (RBF), and Artificial Neural Networks (ANN) in terms of classification accuracy. The outstanding performance of our model in the realm of brain tumor diagnosis signifies a substantial leap forward in the field, holding the promise of delivering more precise and dependable tools for radiologists and healthcare professionals in their pivotal role of identifying and classifying brain tumors using MRI imaging techniques.

Faults Diagnosis for Large-Scale Battery Packs via Texture Analysis on Spatial–Temporal Images Converted From Electrical Behaviors

Battery failures have become the most intractable obstacles undermining the market confidence in applications like electric vehicle and power grid energy storage. This paper aims to fashion a generic diagnosis scheme against the faults in large-scale battery systems. First, a voltmeter array-based anomaly perception mechanism against the electrical behaviors of battery packs is developed. Then, system information on spatial arrangement and temporal dynamics is organically fused and drawn as a kind of pseudo 2D images (P2Is). Afterwards, by analyzing the resultant P2Is with the 2D variational mode decomposition (2D-VMD) and gray level co-occurrence matrix (GLCM), some statistical quantities concerning multi-scale texture features, extracted and refined by the principal component analysis (PCA), are found to have strong indicative associations with battery fault type and fault grade. Finally, relying on the multi-class relevance vector machine (M-RVM), feature evidences are synthesized to detect fault occurrence and give judgements on fault specifics of type and severity. Experimental verifications on a li-ion battery pack with 180 cells suggest that the proposed scheme behaves well in fault type isolating, with an accuracy rate of 97.6% and in fault severity grading, with an accuracy rate of 84.67%.

Multi-scale LBP fusion with the contours from deep CellNNs for texture classification

In texture classification, local binary pattern (LBP) is currently one of the most widely-concerned feature encoding models. Most existing LBP-based texture classification methods are usually limited to single-kind texture features. In fact, an across-domain fusion of LBP features with other features, such as image contours, could be another potential path to promote texture classification. To enhance the feature modeling ability of LBP-based methods, this paper firstly designs a Cellular Neural Network (CellNN) with recurrent convolutions, initially trained by a simplified simulated-annealing algorithm, to extract informative image contours. For better reliability, a new three-channel contour extractor of deep CellNNs (i.e., dCellNNs) is proposed. This extractor contains the initially-trained CellNNs of more than three layers, and it is further optimized by fine-tuning parameters. Moreover, a new weighted-base algorithm is designed to fulfill the fusion of the multi-scale texture features by LBPs and the contour features by dCellNNs to enhance feature representation. Finally, these enhanced features are concatenated together to generate the final multi-scale features of given texture image. On texture datasets KTH, Brodatz, OTC12 and UIUC, experiment results verify that the across-domain fusion of multi-scale LBPs and dCellNNs is efficient in capturing & enhancing texture features. With moderate feature dimensionality and computational costs, it could improve texture classification, acquiring an obvious accuracy increase on previous state-of-the-art ones, e.g., a rise of 2.58% on KTH-TIPS2b, a rise of 3.11% on Brodatz, a rise of 0.71% on OTC12 and a rise of 0.42% on UIUC.

Semantic prior-driven fused contextual transformation network for image inpainting

Recent advances in image inpainting have achieved impressive performance for generating plausible visual details on small regular image defects or simple backgrounds. However, current solution suffers from the lack of semantic priors for the image and the inability to deduce the image content from distant background, leading to distorted structures and artifacts in the results when inpainting large random irregular complicated images. To address these problems, a semantic prior-driven fused contextual transformation network for image inpainting is proposed as a promise solution. First, the semantic prior generator is put forward to map the semantic features of ground truth images and the low-level features of broken images to semantic priors. Subsequently, an image split-transform-aggregated strategy, named fusion context transformation block, is presented to infer rich multi-scale remote texture features and thus to improve the restored image finesse. Thereafter, an aggregated semantic attention-aware module, consisting of spatially adaptive normalization and enhanced spatial attention is designed to aggregate semantic priors and multi-scale texture features into the decoder to restore reasonable structure. Finally, the mask guided discriminator is developed to effectively discriminate between real and false pixels in the output image to improve the capability of the discriminator and hence to reduce the probability of artifacts containing in the output image. Comprehensive experimental results on CelebA-HQ, Paris Street View, and Places2 datasets demonstrate the superiority of the proposed network over the state-of-the-arts, whose PSNR, SSIM and MAE are improved about 20 %, 12.6 %, and 42 % gains, respectively.

Experimental evaluation of unsupervised image retrieval application using hybrid feature extraction by integrating deep learning and handcrafted techniques

Abstract Content Based Image Retrieval is ever growing technology for many applications such as medical, remote sensing, social media search engines and surveillance monitoring, etc,. Representing image content with appropriate features is tedious task using traditional low level feature extraction methods. Deep learning models achieved high precision in classification and object detection algorithms by extracting automated high level feature extraction process. This paper proposed a hybrid feature extraction technique by combining the high level features and low level features to improve the robustness of the feature vector. The proposed model used pre-trained Googlenet model as feature extractor and combined with Gabor multiscale texture features. The final feature vector will be used for retrieving the relevant image data from the large scale image dataset. It has achieved the precision of 91 percent which shows better than state of art methods.

Predicting necessity of daily online adaptive replanning based on wavelet image features for MRI guided adaptive radiation therapy

PurposeOnline adaptive replanning (OLAR) is generally labor-intensive and time-consuming during MRI-guided adaptive radiation therapy (MRgART). This work aims to develop a method to determine OLAR necessity during MRgART. MethodsA machine learning classifier was developed to predict OLAR necessity based on wavelet multiscale texture features extracted from daily MRIs and was trained and tested with data from 119 daily MRI datasets acquired during MRgART for 24 pancreatic cancer patients treated on a 1.5 T MR-Linac. Spearman correlations, interclass correlation (ICC), coefficient of variance (COV), t-test (p < 0.05), self-organized map (SOM) and maximum stable extremal region (MSER) algorithm were used to determine candidate features, which were used to build the prediction models using Bayesian classifiers. The model performance was judged using the AUC of the ROC curve. ResultsSpearman correlation identified 123 features that were not redundant (r < 0.9). Of them 82 showed high ICC for repositioning > 0.6, 67 had a COV greater than 9% for OLAR. Among the 38 features passed the t-test, 25 passed the SOM and 12 passed the MSER. These final 12 features were used to build the classifier model. The combination of 2–3 features at a time was used to build the classifier models. The best performing model was a 3-feature combination, which can predict OLAR necessity with a CV-AUC of 0.98. ConclusionsA machine learning classifier model based on the wavelet features extracted from daily MRI for pancreatic cancer was developed to automatically and objectively determine if OLAR is necessary for a treatment fraction avoiding unnecessary effort during MRgART.

Multi-Scale LBP Texture Feature Learning Network for Remote Sensing Interpretation of Land Desertification

Land desertification is a major challenge to global sustainable development. Therefore, the timely and accurate monitoring of the land desertification status can provide scientific decision support for desertification control. The existing automatic interpretation methods are affected by factors such as “same spectrum different matter”, “different spectrum same object”, staggered distribution of desertification areas, and wide ranges of ground objects. We propose an automatic interpretation method for the remote sensing of land desertification that incorporates multi-scale local binary pattern (MSLBP) and spectral features based on the above issues. First, a multi-scale convolutional LBP feature extraction network is designed to obtain the spatial texture features of remote sensing images and fuse them with spectral features to enhance the feature representation capability of the model. Then, considering the continuity of the distribution of the same kind of ground objects in local space, we designed an adaptive median filtering method to process the probability map of the extreme learning machine (ELM) classifier output to improve the classification accuracy. Four typical datasets were developed using GF-1 multispectral imagery with the Horqin Left Wing Rear Banner as the study area. Experimental results on four datasets show that the proposed method solves the problem of ill classification and omission in classifying the remote sensing images of desertification, effectively suppresses the effects of “homospectrum” and “heterospectrum”, and significantly improves the accuracy of the remote sensing interpretation of land desertification.

Optimum Feature and Classifier Selection for Accurate Urban Land Use/Cover Mapping from Very High Resolution Satellite Imagery

Feature selection to reduce redundancies for efficient classification is necessary but usually time consuming and challenging. This paper proposed a comprehensive analysis for optimum feature selection and the most efficient classifier for accurate urban area mapping. To this end, 136 multiscale textural features alongside a panchromatic band were initially extracted from WorldView-2, GeoEye-3, and QuickBird satellite images. The wrapper-based and filter-based feature selection were implemented to optimally select the best ten percent of the primary features from the initial feature set. Then, machine leaning algorithms such as artificial neural network (ANN), support vector machine (SVM), and random forest (RF) classifiers were utilized to evaluate the efficiency of these selected features and select the most efficient classifier. The achieved optimum feature set was validated using two other images of WorldView-3 and Pleiades. The experiments revealed that RF, particle swarm optimization (PSO), and neighborhood component analysis (NCA) resulted in the most efficient classifier and wrapper-based and filter-based methods, respectively. While ANN and SVM’s process time depended on the number of input features, RF was significantly resistant to the criterion. Dissimilarity, contrast, and correlation features played the greatest contributing role in the classification performance among the textural features used in this study. These trials showed that the feature number could be reduced optimally to 14 from 137; these optimally selected features, alongside the RF classifier, can produce an F1-measure of about 0.90 for different images from five very high resolution satellite sensors for various urban geographical landscapes. These results successfully achieve our goal of assisting users by eliminating the task of optimal feature selection and classifier, thereby increasing the efficiency of urban land use/cover classification from very high resolution images. This optimal feature selection can also significantly reduce the high computational load of the feature-engineering phase in the machine and deep learning approaches.

Single-Image Super-Resolution Neural Network via Hybrid Multi-Scale Features

In this paper, we propose an end-to-end single-image super-resolution neural network by leveraging hybrid multi-scale features of images. Different from most existing convolutional neural network (CNN) based solutions, our proposed network depends on the observation that image features extracted by CNN contain hybrid multi-scale features: both multi-scale local texture features and global structural features. By effectively exploiting these multi-scale and local-global features, our network involves far fewer parameters, leading to a large decrease in memory usage and computation during inference. Our network benefits from three key modules: (1) an efficient and lightweight feature extraction module (EFblock); (2) a hybrid multi-scale feature enhancement module (HMblock); and (3) a reconstruction–restoration module (DRblock). Experiments on five popular benchmarks demonstrate that our super-resolution approach achieves better performance with fewer parameters and less memory consumption, compared to more than 20 SOTAs. In summary, we propose a novel multi-scale super-resolution neural network (HMSF), which is more lightweight, has fewer parameters, and requires less execution time, but has better performance than the state-of-the-art methods. Compared to SOTAs, this method is more practical and better suited to run on constrained devices, such as PCs and mobile devices, without the need for a high-performance server.

An Adaptive Learning Model for Multiscale Texture Features in Polyp Classification via Computed Tomographic Colonography.

Objective: As an effective lesion heterogeneity depiction, texture information extracted from computed tomography has become increasingly important in polyp classification. However, variation and redundancy among multiple texture descriptors render a challenging task of integrating them into a general characterization. Considering these two problems, this work proposes an adaptive learning model to integrate multi-scale texture features. Methods: To mitigate feature variation, the whole feature set is geometrically split into several independent subsets that are ranked by a learning evaluation measure after preliminary classifications. To reduce feature redundancy, a bottom-up hierarchical learning framework is proposed to ensure monotonic increase of classification performance while integrating these ranked sets selectively. Two types of classifiers, traditional (random forest + support vector machine)- and convolutional neural network (CNN)-based, are employed to perform the polyp classification under the proposed framework with extended Haralick measures and gray-level co-occurrence matrix (GLCM) as inputs, respectively. Experimental results are based on a retrospective dataset of 63 polyp masses (defined as greater than 3 cm in largest diameter), including 32 adenocarcinomas and 31 benign adenomas, from adult patients undergoing first-time computed tomography colonography and who had corresponding histopathology of the detected masses. Results: We evaluate the performance of the proposed models by the area under the curve (AUC) of the receiver operating characteristic curve. The proposed models show encouraging performances of an AUC score of 0.925 with the traditional classification method and an AUC score of 0.902 with CNN. The proposed adaptive learning framework significantly outperforms nine well-established classification methods, including six traditional methods and three deep learning ones with a large margin. Conclusions: The proposed adaptive learning model can combat the challenges of feature variation through a multiscale grouping of feature inputs, and the feature redundancy through a hierarchal sorting of these feature groups. The improved classification performance against comparative models demonstrated the feasibility and utility of this adaptive learning procedure for feature integration.

Multiscale Structure and Texture Feature Fusion for Image Inpainting

In order to achieve interaction between structure and texture information in generative adversarial image inpainting networks and improve the semantic veracity of the restored images, unlike the original two-stage inpainting ideas where texture and structure are restored separately, this paper constructs a multi-scale fusion approach to image generation, which embeds images into two collaborative subtasks, that is, structure generation and texture synthesis under structural constraints. We also introduce a self-attention mechanism into the partial convolution of the encoder to enhance the long range contextual information acquisition of the model in image inpainting, and design a multi-scale fusion network to fuse the generated structure and texture feature, so that the structure and texture information can be reused for reconstruction, perception and style loss compensation, thus enabling the fused images to achieve global consistency. In the training phase, feature matching loss are introduced to enhance the image in terms of structural generation plausibility. Finally, through comparison experiments with other inpainting networks on the CelebA, Paris StreetView and Places2 datasets, it is demonstrated that our method constructed in this paper has better objective evaluation metrics, more effective inpainting of structural and texture information of corrupted images and better image inpainting performance.

Assessing Surface Texture Features of Asphalt Pavement Based on Three-Dimensional Laser Scanning Technology

Pavement surface texture features are one of key factors affecting the skid resistance of pavement. In this study, a set of stable and reliable texture measurement equipment was firstly assembled by using the linear laser ranging sensor, control system and data acquisition system. Secondly, the equipment was calibrated, and the superposition error of sensor and control system was tested by making a standard gauge block. Thirdly, four different kinds of asphalt mixture were designed, and their surface texture features were obtained by leveraging a three-dimensional laser scanner. Therefore, the surface texture features were characterized as one-dimensional profile features and three-dimensional surface features. At the end of this study, a multi-scale texture feature characterization method was proposed. Results demonstrate that the measurement accuracy of the laser scanning system in the x-axis direction can be controlled ranging from −0.01 mm to 0.01 mm, the resolution in the XY plane is 0.05 mm, and the reconstructed surface model of surface texture features can achieve a good visualization effect. They also show that the root mean square deviation of surface profiles of different asphalt pavements fluctuates greatly, which is mainly affected by the nominal particle size of asphalt mixture and the proportion of coarse aggregate, and the non-uniformity of pavement texture distribution makes it difficult to characterize the roughness of asphalt pavement effectively by a single pavement surface profile. This study proposed a texture section method to describe the 3D distribution of road surface texture at different depths. The macrotexture of the road surface gradually changes from sparse to dense starting from the shallow layer. The actual asphalt pavement texture can be characterized by a simplified combination model of “cone + sphere + column”. By calculating the surface area distribution of macro and microtextures of different asphalt pavements, it was concluded that the surface area of asphalt pavement under micro scale is about 1.8–2.2 times of the cutting area, and the surface area of macrotexture is about 1.4 times of the cutting area. Moreover, this study proposed texture distribution density to characterize the roughness of asphalt pavement texture at different scales. The SMA index can represent the macroscopic structure level of different asphalt pavements to a certain extent, and the SMI index can well represent the friction level of different asphalt pavements.

Automatically Determining Necessity of Online Adaptive Replanning Based on MRI Wavelet Multiscale Texture Features for MRI-Guided Adaptive Radiation Therapy

MRI-guided adaptive radiation therapy (MRgART), particularly daily online adaptive replanning (OLAR) as enabled with MR-LINAC, can substantially improve radiation therapy delivery. However, OLAR is labor-intensive and time-consuming. OLAR is unnecessary for all treatment fractions and being able to objectively and automatically determine when it is needed is desirable. In this study, we investigate whether wavelet multiscale texture features extracted from daily MRI can be used to automatically and objectively identify when OLAR is needed using a machine learning approach.The method was developed and demonstrated using 70 daily motion average MRI sets acquired on a 1.5T MR-LINAC using a balanced turbo field echo sequence during MRgART for 14 pancreatic cancer patients each treated with 5 fractions. For each daily MRI set, OLAR and repositioning (i.e., adapt to position) plans were created and judged per clinical dose-volume constrains by experienced MRL physicists. Multiscale texture features were extracted from the region enclosed by 50 and 100% reference isodose surfaces for different wavelet decomposition levels. Spearman correlations were utilized to rule-out redundant features. Inter-class correlation (ICC), coefficient of variance (COV), and t-test (P < 0.05) were used to determine significant features. Mahalonobis distance classifier with a leave one out cross-validation method was used to develop prediction models based on the identified features and their combinations. The performance of the model to predict whether OLAR is necessary for a given daily MRI set was measured using the AUC of the ROC curve.Spearman correlation identified 123 features from different wavelet decomposition levels that were not redundant (r < 0.9). Of these features, 82 showed high ICC for repositioning, 67 had a COV greater than 9% for OLAR and lower than 5% for repositioning. In addition, 38 of the 67 features passed the t-test with P value < 0.05. Best performing model was a 3-feature combination that can predict OLAR necessity with an AUC of 0.95. It was found that misclassification was attributed to some cases with very close dosimetric gain between repositioning and OLAR and hence a lower discrimination ability for this subgroup.The necessity of OLAR can be automatically determined using a machine learning model based on the wavelet multiscale texture features from the daily MRI immediately after its acquisition, avoiding unnecessary effort during MRgART.

Medical Image Segmentation Algorithm Based on Multi-scale Color Wavelet Texture

In order to improve the segmentation accuracy of brain tumor magnetic resonance medical image, a segmentation method of brain tumor magnetic resonance medical image based on multi-scale color wavelet texture features is proposed. The segmentation model of brain tumor magnetic resonance medical image is established, and the motion damage information of brain tumor magnetic resonance medical image is adaptively fused in the ultrasound imaging environment. The medical image information is enhanced by using the motion skeletal muscle block matching technology. According to the suspicious point feature matching method of brain tumor, the fusion detection and processing of brain tumor magnetic resonance medical image are carried out. The multi-scale color wavelet texture feature detection method is used to extract the image features of brain tumor MRI points, and the CT bright spot features are used to analyze the features of brain tumor MRI medical images. Combined with the adaptive neural network training method, the automatic detection of brain tumor magnetic resonance medical image is completed, and the suspected brain tumor points are extracted, so as to realize the segmentation of brain tumor magnetic resonance medical image. Simulation results show that the proposed method can effectively improve the segmentation accuracy of brain tumor MRI medical image, and has high resolution and accuracy for suspicious brain tumor detection.

An ultrasound image-based deep multi-scale texture network for liver fibrosis grading in patients with chronic HBV infection.

Background & AimsThe evaluation of the stage of liver fibrosis is essential in patients with chronic liver disease. However, due to the low quality of ultrasound images, the non‐invasive diagnosis of liver fibrosis based on ultrasound images is still an outstanding question. This study aimed to investigate the diagnostic accuracy of a deep learning‐based method in ultrasound images for liver fibrosis staging in multicentre patients.MethodsIn this study, we proposed a novel deep learning‐based approach, named multi‐scale texture network (MSTNet), to assess liver fibrosis, which extracted multi‐scale texture features from constructed image pyramid patches. Its diagnostic accuracy was investigated by comparing it with APRI, FIB‐4, Forns and sonographers. Data of 508 patients who underwent liver biopsy were included from 4 hospitals. The area‐under‐the ROC curve (AUC) was determined by receiver operating characteristics (ROC) curves for significant fibrosis (≥F2) and cirrhosis (F4).ResultsThe AUCs (95% confidence interval) of MSTNet were 0.92 (0.87‐0.96) for ≥F2 and 0.89 (0.83‐0.95) for F4 on the validation group, which significantly outperformed APRI, FIB‐4 and Forns. The sensitivity and specificity of MSTNet (85.1% (74.5%‐92.0%) and 87.6% (78.0%‐93.6%)) were better than those of three sonographers in assessing ≥F2.ConclusionsThe proposed MSTNet is a promising ultrasound image‐based method for the non‐invasive grading of liver fibrosis in patients with chronic HBV infection.

Using remote sensing to identify soil types based on multiscale image texture features

Studying the spatial distribution of soil types is an important academic and practical issue in agriculture. With the rapid development of remote sensing technology, the role of image texture as an auxiliary variable in remote sensing identification of objects has increased. It is of great importance to ascertain the optimal window size for extracting texture features and the multiscale fusion of texture feature parameters under the optimal window for different soil types. To reach this goal, soil types in a typical area of the Jiaodong Peninsula were selected as the subject investigated, homogeneity and entropy were selected as the two texture feature parameters, and the ability to identify the different soil types based on the textural features was systematically analyzed by using Landsat 8 remote sensing images. Moreover, the optimal window sizes for extracting texture features were determined, and the role of multiscale textural features in the classification of the soil types was also evaluated. The results show that the accuracy of classification significantly increased with the addition of textural features. The optimal single-scale window sizes for the homogeneity and entropy feature parameters were 19 × 19 and 21 × 21, respectively. The fusion of multiscale textural features further improved the classification accuracy. The optimal multiscale window sizes for the homogeneity were 7 × 7, 13 × 13, 19 × 19 and 21 × 21 and those for entropy were 5 × 5, 15 × 15, 21 × 21 and 23 × 23. Therefore, the method of using texture information in remote sensing images as auxiliary variables in digital soil mapping was feasible. The method of multiscale fusion of texture features, which resulted in greater classification accuracy, was better than that of single-scale window. These conclusions could play an important guiding role in soil digital mapping with remote sensing.

Improved Estimation of Winter Wheat Aboveground Biomass Using Multiscale Textures Extracted from UAV-Based Digital Images and Hyperspectral Feature Analysis

Rapid and accurate crop aboveground biomass estimation is beneficial for high-throughput phenotyping and site-specific field management. This study explored the utility of high-definition digital images acquired by a low-flying unmanned aerial vehicle (UAV) and ground-based hyperspectral data for improved estimates of winter wheat biomass. To extract fine textures for characterizing the variations in winter wheat canopy structure during growing seasons, we proposed a multiscale texture extraction method (Multiscale_Gabor_GLCM) that took advantages of multiscale Gabor transformation and gray-level co-occurrency matrix (GLCM) analysis. Narrowband normalized difference vegetation indices (NDVIs) involving all possible two-band combinations and continuum removal of red-edge spectra (SpeCR) were also extracted for biomass estimation. Subsequently, non-parametric linear (i.e., partial least squares regression, PLSR) and nonlinear regression (i.e., least squares support vector machine, LSSVM) analyses were conducted using the extracted spectral features, multiscale textural features and combinations thereof. The visualization technique of LSSVM was utilized to select the multiscale textures that contributed most to the biomass estimation for the first time. Compared with the best-performing NDVI (1193, 1222 nm), the SpeCR yielded higher coefficient of determination (R2), lower root mean square error (RMSE), and lower mean absolute error (MAE) for winter wheat biomass estimation and significantly alleviated the saturation problem after biomass exceeded 800 g/m2. The predictive performance of the PLSR and LSSVM regression models based on SpeCR decreased with increasing bandwidths, especially at bandwidths larger than 11 nm. Both the PLSR and LSSVM regression models based on the multiscale textures produced higher accuracies than those based on the single-scale GLCM-based textures. According to the evaluation of variable importance, the texture metrics “Mean” from different scales were determined as the most influential to winter wheat biomass. Using just 10 multiscale textures largely improved predictive performance over using all textures and achieved an accuracy comparable with using SpeCR. The LSSVM regression model based on the combination of the selected multiscale textures, and SpeCR with a bandwidth of 9 nm produced the highest estimation accuracy with R2val = 0.87, RMSEval = 119.76 g/m2, and MAEval = 91.61 g/m2. However, the combination did not significantly improve the estimation accuracy, compared to the use of SpeCR or multiscale textures only. The accuracy of the biomass predicted by the LSSVM regression models was higher than the results of the PLSR models, which demonstrated LSSVM was a potential candidate to characterize winter wheat biomass during multiple growth stages. The study suggests that multiscale textures derived from high-definition UAV-based digital images are competitive with hyperspectral features in predicting winter wheat biomass.

Texture feature extraction and optimization of facial expression based on weakly supervised clustering

In order to improve the recognition rate of weak annotation data in facial expression recognition task, this paper proposes a multi-scale and multi-region vector triangle texture feature extraction scheme based on weakly supervised clustering algorithm. According to the information gain rate of extracted features, combined with threshold selection and random dropout strategy, the best selection of vector triangle texture feature scale is explored, and the feature space is optimized under the premise of sufficient feature space information, the reduction of feature space is realized and the information redundancy is reduced. For the positive and negative expression units, the facial expression images in the data set are divided into two categories. The positive and negative facial expressions are taken to form the same kind of samples, the positive and negative facial expressions are taken to form the positive and negative samples, and the annotation labels are taken to form the weak annotation labels. The experimental results show that the best recognition rate of the proposed scheme is 84.1%, which is 5.8% higher than the unoptimized texture feature scheme.

An Independent and Automatic Quality Assurance Tool for Auto-Segmentation

Deep-learning based auto-segmentation has been shown to be efficient and accurate or consistent for radiotherapy planning. However, in current practice, manual review of auto-segmented contours before their clinical use is necessary and can be labor intensive and time consuming. This study aims to develop an independent and automatic contour quality assurance (QA) tool that can quickly identify inaccurate or inconsistent contour slices from auto-segmented contours, eliminating the need of manual QA process. The automatic contour QA tool starts from building a model, which includes: (1) pre-processing images, (2) extracting quantitative geometric and multi-scale texture features from the test contours and the feature changes between 4-mm inner/outer shells and a core region of the test contours on a slice-by-slice basis, (3) selecting a subset of organ-specific discriminative features that maximizes the model’s predictive performance using a recursive feature elimination method, and (4) determining the best performing model among 12 different machine learning-based supervised classification models. The process was tested using an MRI head and neck dataset. Contours of the parotid glands and submandibular glands on T2-weighted non-contrast MRIs from 31 patients were utilized, where 21 were used for training and 10 for testing. During model training, manually-delineated contours for each training case were utilized as accurate samples. The inaccurate samples were collected from a set of deformable image registration (DIR) propagated contours using a labeling criterion of mean distance to agreement ≥ 2 mm or Hausdorff distance ≥ 8mm as compared to ground truth contours. For the 10 MRIs in the test set, two sets of contours using a deep-learning based auto-segmentation software and DIR propagation were used to evaluate the model performance. All the contour slices were labeled using the same labeling criterion and manually checked to avoid mislabeling. The classification accuracy was used to measure a model’s performance. For each contour slice, a total of 437 features were extracted. At the training stage, 58, 41, 72, and 56 features were automatically selected for left and right parotid gland and submandibular glands contours, respectively. The best model had an accuracy of 89% for the four organs. The time required to evaluate one patient dataset is less than 30 seconds using a workstation equipped with a i7-6700 CPU. The developed automatic contour QA tool can quickly identify accurate and inaccurate contour slices with high accuracy from auto-segmented contours in head and neck. This tool is being integrated into a deep learning-based auto-segmentation package, to eventually develop a fully automatic and robust auto-segmentation pipeline for radiotherapy planning and delivery.