Error Map Research Articles

Evaluating principal component analysis models for representing anatomical changes in head and neck radiotherapy.

Background and purposeAnatomical changes during radiotherapy pose a challenge to robustness of plans. Principal component analysis (PCA) is commonly used to model such changes. We propose a toolbox to evaluate how closely a given PCA model can represent actual deformations seen in the patient and highlight regions where the model struggles to capture these changes.Materials and methodsWe propose to calculate a residual error map from the difference between an actual displacement vector field (DVF) and the closest DVF that the PCA model can produce. This was done by taking the inner product of the DVF with the PCA components from the model. As a global measure of error, the 90th percentile of the residual errors ) across the whole scan was used. As proof of principle, we demonstrated this approach on both patient-specific cases and a population-based PCA in head and neck (H&N) cancer patients. These models were created using deformation data from deformable registrations between the planning computed tomography and cone-beam computed tomography (CBCTs), and were evaluated against DVFs from registrations of CBCTs not used to create the model.ResultsFor our example cases, the oropharyngeal and the nasal cavity regions showed the largest local residual error, indicating the PCA models struggle to predict deformations seen in these regions. ranged from 0.4 mm to 6.3 mm across the different models.ConclusionsA method to quantitatively evaluate how well PCA models represent observed anatomical changes was proposed. We demonstrated our approach on H&N PCA models, but it can be applied to other sites.

Breakdown of light transport models in photonic scattering slabs with strong absorption and anisotropy

The radiative transfer equation (RTE) models the transport of light inside photonic scattering samples such as paint, foam and tissue. Analytic approximations to solve the RTE fail for samples with strong absorption and dominant anisotropic scattering and predict unphysical negative energy densities and the diffuse flux in the wrong direction. Here we fully characterize the unphysical regions of three popular approximations to the RTE for a slab, namely the $P_1$ approximation (or diffusion approximation), the $P_3$ approximation, and a popular modification to $P_3$ that corrects the forward scattering in the approximation. We find that the delta function correction to $P_3$ eliminates the unphysical range in the forward scattering. In addition, we compare the predictions of these analytical methods to exact Monte Carlo simulations for the physical and unphysical regions. We present maps of relative errors for the albedo and the anisotropy of the scatterers for a realistic index contrast typical of a polymer slab in air and optical thickness. The relative error maps provide a guideline for the accuracy of the analytical methods to interpret experiments on light transport in photonic scattering slabs. Our results show that the $P_1$ approximation is significantly inaccurate to extract transport parameters unless the sample scatters purely isotropic and elastic. The $P_3$ approximation exceeds $P_1$ in terms of accuracy in its physical range for moderate absorption, and the $P_3$ with the delta function correction is the most accurate approximation considered here for the forward direction.

Precision analysis of series-parallel machine tool based on hierarchy motion function units

This paper takes the motion units of the series-parallel machine tool (SPMT) as the basic research objects. Combined with the meta-action theory, the FMA-MBS structure model is proposed to describe the spatial structure relationships and motion transfer relationships of SPMT. Based on the simplified model, the kinematics characteristics of the machine tool are gradually researched from four layers: joint layer, limb layer, subsystem layer, and the whole machine. According to the twist system of limbs and moving platform, the parallel motion system can be simplified into a virtual motion chain with specific constraints. Then, the kinematics model of SPMT is constructed in twist exponential form. By introducing two virtual degrees of freedom (DOFs), the unified expressions of active and passive joint errors are obtained. All motion systems error models are derived based on three basic error terms of the meta-action layer. Based on the error mapping relationships, the precision of the machine tool can be calculated and evaluated. In order to illustrate the applicability and effectiveness of the proposed method, a sample library of common SPMTs and a case study of a 5-axis series-parallel machining center are provided.

Definition of an error map for DoFP polarimetric images and its application to retardance calibration.

With the recent development of division of focal plane (DoFP) polarization sensors, it is possible to perform polarimetric analysis of a scene with a reduced number of acquisitions. One drawback of these sensors is that polarization estimation can be perturbed by the spatial variations of the scene. We thus propose a method to compute a map that indicates where polarization estimation can be trusted in the image. It is based on two criteria: the consistency between the intensity measurements inside a super-pixel and the detection of spatial intensity variations. We design both criteria so that a constant false alarm rate can be set. We demonstrate the benefit of this method to improve the precision of dynamic retardance calibration of DoFP-based full Stokes imaging systems.

Mapping and Analysis of Standard Indonesian Pronunciation Errors by Using the Bigram Method

Indonesian language is increasingly being ignored, even the mass media often find the use of non-standard language, so there is a uniformity in the use of words that often appear in scientific articles, especially those Indonesian. The uniformity of Indonesian pronunciation certainly confuses the general public, for example: television news viewers and radio listeners, to distinguish between standard and non-standard forms. The non-uniformity of Indonesian pronunciation often occurs in official situations such as official speeches or presentations. Based on this phenomenon, this study aims to conduct an analysis and mapping study of standard pronunciation errors that arise in several public services and develops application tools as automatically tool that can detect errors from the use of standard words obtained from voice recording results in the form of vowel errors, diphthongs and consonants by implementing the bigram method. This research was conducted by collecting voice recordings of conversations, speeches, public speaking and other recordings deemed necessary, then continued with standard pronunciation errors analysis and mapping of these errors using a tool that applied the Bigram technique. Based on statistical information obtained from the detection and correction process that has been carried out, it can be obtained a mapping of the error types in the use of Indonesian standard that often occur in public service environment. The results of the mapping and statistical information obtained are used as the basis for developing programs that aim to improve the quality of the use Indonesian standard in the form of FGDs, socialization and counseling. The study result indicate the standard of pronunciation errors that occur in various circles including presenters and lecturers, then the results of data analysis are described in the form of tables and diagrams so it is seen the percentage of errors that occur

Recaptured image forensics based on local ternary count of high order prediction error

With the advancement in image display technology and the widespread use of various image acquisition devices, recapturing high-quality images from high-fidelity LCD (liquid crystal display) screens becomes very convenient and easy. These recaptured images pose serious threats on current image forensic technology and intelligent recognition systems. In order to solve these security problems, we propose a recaptured image forensics method based on local ternary count (LTC) of high order prediction error. Specifically, given an image, the second-order difference maps on different directions are calculated in HSV color space for the original image and its average pooled version. Then, we calculate the prediction error maps from these second-order difference maps through a linear predictor. Finally, the normalized LTC histograms of all the downsampled versions of the obtained prediction error maps are calculated as features for training and testing. Single database experiments demonstrate that our proposed method performs very similar with the state-of-the-art traditional method on the most difficult-to-detect ICL-COMMSP database with obvious less execution-time and lower feature dimension. Meanwhile, it performs very close to the existing well-performed deep learning based method on all the databases, which verifies the effectiveness of our proposed features. Besides, mixed database experiments verify the superiority on generalization ability of our proposed method. Moreover, it is robust to JPEG compression.

Spectrally Consistent Mean Dynamic Topography by Combining Mean Sea Surface and Global Geopotential Model Through a Least Squares-Based Approach

The filtering procedure is usually mandatory for modeling mean dynamic topography (MDT) when a geodetic approach based on the Mean Sea Surface (MSS) and the Global Geopotential Model (GGM) is used. This is due to the inconsistent spectral contents between MSS and GGM. However, traditional isotropic filtering algorithms (e.g., Gaussian filter) consider neither the MDT locations nor their azimuth when smoothing the signal within the filtering radius. Hence, the isotropic filtering will attenuate the MDT signal near the current and filter the current signal into the surrounding ocean, which may lead to signal contamination and distortion. In this study, we set up a least squares-based (LS) approach to model MDT signal from the altimeter-derived MSS and geoid height using spherical harmonics from GGMs, where MDT is parameterized by Lagrange Basis Functions (LBFs). The design matrix is segmentally established, considering the error information of GGM in various spectral bands. Numerical experiments in the Gulf Stream show that applications of full error variance-covariance matrix or only diagonal error variance of GGM may have marginal effects on the MDT modeling. The MDT computed from this LS-based approach using the latest releases of Gravity Field and Steady-state Ocean Circulation Explorer (GOCE) geoid models, i.e., GO_CONS_GCF_2_DIR_R6 and Gravity Observation Combination 06s model (GOCO06s), have the best agreement with the comparison data, especially near the current region. Deduced geostrophic velocities based on the MDT solutions show that the LS-based approach recovers the current signal better than the Gaussian filtering by 1.8 cm/s. Estimated error map illustrates that errors are more concentrated near the coastal region.

Multi-modal image feature fusion-based PM2.5 concentration estimation

PM2.5 can be suspended in the air for a long time, so it has great harm to human health and atmospheric environment quality. Real-time and reliable estimation of PM2.5 concentration is of great significance. This paper proposes a novel multi-modal image information fusion-based PM2.5 concentration estimator. Specifically, in the depth domain, we estimate PM2.5 concentration by calculating the depth error between the PM2.5 image and its corresponding dehazing result. In the texture domain, we measure PM2.5 concentration using the following two modules: (1) The PM2.5 concentration is represented by calculating the entropy difference between the PM2.5 image and its dehazing image; (2) We first compute the error map between the PM2.5 image and its dehazing image, and then calculate the color similarity between the PM2.5 image and the obtained error map to measure PM2.5 concentration. Finally, we use support vector regression to perform regression learning on the above-mentioned multi-modal features learned from the depth domain and texture domain to obtain the final PM2.5 concentration estimator. Experiments show that the proposed method can measure PM2.5 concentration more efficiently than the existing PM2.5 concentration estimation algorithms.

Spatial-temporal Analysis of Soybean Productivity Using Geostatistical Methods

To have information about the soybean productivity over the crop years is essential to define strategies to increase profits and reduce costs and most important to reduce environmental impacts. One form of monitoring is the use of Geostatistical methods, which allow us to obtain maps with more accurate predictions. In this paper, an area of 127.16 ha was studied during six crop years between 2012/2013 and 2017/2018. We found that productivity values vary between crop years, mainly due to uncontrollable climatic factors. The removal of influential points caused changes in the predicted values showed in the maps, and the use of scaled semivariograms allowed us to obtain similar maps to those obtained considering the model without influential points, then there was no need to exclude observations. The use of a model with replicates helped to identify regions where productivity was lower. The use of explanatory variables allowed us to elaborate a more accurate thematic map in the 2017/2018 crop year, which was well evidenced by the prediction standard error map.

Variable Gain Prescribed Performance Control for Dynamic Positioning of Ships with Positioning Error Constraints

In this paper, a variable gain prescribed performance control law is proposed for dynamic positioning (DP) of ships with positioning error constraints, input saturation and unknown external disturbances. The error performance index functions are designed to preset the prescribed performance bounds and the error mapping functions are constructed to incorporate the prescribed performance bounds into the DP control design. The variable gain technique is used to limit the output amplitude of the control law to avoid input saturation of the system by dynamically adjusting the control gain of the DP control law according to the positioning errors, and the error mapping function replaces the positioning error as a recursive sliding-mode surface to realize the prescribed performance control of the system and guarantee the stability of the closed-loop system with variable control gains. It has been proved that the proposed DP control law can make the uniformly ultimately boundedness of all signals in the DP closed-loop control system. The numerical simulation results illustrate that the proposed control law can make the ship’s position and heading maintain at the desired value with positioning error constraints, enhance the non-fragility of the DP control law to the perturbation of system’s parameters and improve the system’s rejection ability to external disturbances.

Real-space RG, error correction and Petz map

There are two parts to this work: first, we study the error correction properties of the real-space renormalization group (RG). The long-distance operators are the (approximately) correctable operators encoded in the physical algebra of short-distance operators. This is closely related to modeling the holographic map as a quantum error correction code. As opposed to holography, the real-space RG of a many-body quantum system does not have the complementary recovery property. We discuss the role of large N and a large gap in the spectrum of operators in the emergence of complementary recovery.Second, we study the operator algebra exact quantum error correction for any von Neumann algebra. We show that similar to the finite dimensional case, for any error map in between von Neumann algebras the Petz dual of the error map is a recovery map if the inclusion of the correctable subalgebra of operators has finite index.

ToF and Stereo Data Fusion Using Dynamic Search Range Stereo Matching

Time-of-Flight (ToF) sensors and stereo vision systems are both widely used for capturing depth data. They have some complementary strengths and limitations, which have been exploited in prior research to produce more accurate depth maps by fusing data from the two sources. However, among these diverse data fusion approaches, none of them provides an end-to-end neural network solution. In this work, we propose the first end-to-end ToF and stereo data fusion network using the coarse-to-fine matching framework, where the prior of ToF depth is integrated into the stereo matching process by constraining the search range of stereo matching within an interval around the ToF camera depth measurement. We adopt a dynamic search range for each pixel according to an estimated ToF error map, which is more efficient and effective than a constant one when handling various errors. The ToF error map is estimated by the ToF error estimator branching out from the stereo matching network. Both ToF error estimation and stereo matching are performed in a joint framework, with the two tasks assisting each other mutually. We also propose an upsampling module to replace the naive bilinear upsampling in the coarse-to-fine stereo matching network, which reduces the error caused by the upsampling. The proposed deep network is trained end-to-end on synthetic datasets and generalizable to real-world datasets without further fine-tuning. Experimental results show that our fusion method achieves higher accuracy than either ToF or stereo alone, and outperforms state-of-the-art fusion methods on both synthetic and real data.

Label Augmentation to Improve Generalization of Deep Learning Semantic Segmentation of Laparoscopic Images

Training Deep Neural Networks to solve semantic segmentation is a challenging problem with small-size labeled datasets, leading to overfitting. This is especially problematic in medical images, and in particular, in laparoscopic surgery images. In this context, ground-truth segmentation labels are available only for a small set of images with few patients. Besides, inter-patient variability is very high in practice. Models trained for a specific setup and a set of patients usually performs poorly when deployed in a new environment. This work proposes a new training strategy that improves the generalization accuracy of current state-of-the-art semantic segmentation methods applied to laparoscopic images. Our approach is based on training a discriminator network, which learns to detect segmentation errors, producing a dense segmentation error map. Unlike in adversarial networks, we train the discriminator offline by synthetically altering ground-truth segmentation labels with simple morphological and geometric operations. We then use the discriminator to train a segmentation neural network, by minimizing the discriminator predicted error jointly with a standard segmentation loss. This strategy results in segmentation models that are significantly more accurate when tested in unseen images than those only relying on data augmentation. This technique is very suitable to boost the performance of any state-of-the-art segmentation network and can be combined with other data augmentation strategies. This paper evaluates and validates our proposal by training and testing common state-of-the-art segmentation models in publicly available semantic segmentation datasets, specialized in laparoscopic and endoscopic surgery. The results show that our methods are effective, obtaining a significant improvement in terms of segmentation accuracy, especially in challenging small-size datasets.

Study on Calibration of 3-PR(4R)R Three-Degree-of-Freedom Parallel Mechanism

A kinematic calibration approach is proposed for a triple-planar 3-PR(4R)R parallel mechanism whose attitude error cannot be corrected by motion. First, the kinematic principle of the 3-PR(4R)R parallel mechanism is presented. Then, an error mapping function model is designed based on the mechanism’s closed-loop vector equations. The model reveals the presence of coupling errors in the mechanism, i.e., the coupling of attitude and position errors, and is used to analyze error sensitivity as an indicator of component machining and assembly accuracy. A kinematic inverse solution is used to build a zero-point calibration and a full calibration method. Calibration is performed by a laser tracker-based calibration system, which generates much lower error values and delivers much improved accuracy. Experiments demonstrate that the proposed approach can potentially be used to calibrate parallel mechanisms with non-compensable error terms.

Hyperspectral Anomaly Detection Based on Graph Regularized Variational Autoencoder

Nowadays, deep learning can play an important role in addressing the issue of hyperspectral anomaly detection (HAD). In order to further utilize the spatial information in hyperspectral images (HSIs), an anomaly detection method for HSIs is presented based on graph regularized variational autoencoder (GRVAE). Firstly, the proposed method uses the superpixel segmentation algorithm to segment the hyperspectral image (HSI) and constructs an adjacency matrix to evaluate the similarity between pixels. Secondly, a variational autoencoder is used to reconstruct the spectral vector of the HSI, and meanwhile, the spatial similarity of the image is shared in the feature space through the graph regularization term. Finally, the reconstructed background and the original input are used to obtain the spectral error map, and then the attribute filtering is used to further refine the detection results. Performed on four data sets of abnormal target data with different shapes and different background complexity, the experiments show that the method has promising anomaly detection performance.

Automatic Hemorrhage Detection From Color Doppler Ultrasound Using a Generative Adversarial Network (GAN)-Based Anomaly Detection Method.

Hemorrhage control has been identified as a priority focus area both for civilian and military populations in the United States because exsanguination is the most common cause of preventable death in hemorrhagic injury. Non-compressible torso hemorrhage (NCTH) has high mortality rate and there are currently no broadly available therapies for NCTH outside of a surgical room environment. Novel therapies, which include High Intensity Focused Ultrasound (HIFU) have emerged as promising methods for hemorrhage control as they can non-invasively cauterize bleeding tissue deep within the body without injuring uninvolved regions. A major challenge in the application of HIFU with color Doppler US guidance is the interpretation and optimization of the blood flow images in real-time to identify the hemorrhagic focus. Today, this task requires an expert sonographer, limiting the utility of this therapy in non-clinical environments. In this work, we investigated the feasibility of an automated hemorrhage detection method using a Generative Adversarial Network (GAN) for anomaly detection that learns a manifold of normal blood flow variability and subsequently identifies anomalous flow patterns that fall outside the learned manifold. As an initial feasibility study, we collected ultrasound color Doppler images of femoral arteries in an animal model of vascular injury (N = 11 pigs). Velocity information of the blood flow were extracted from the color Doppler images that were used for training and testing the anomaly detection network. Normotensive images from 8 pigs were used for training, and testing was performed on normotensive, immediately after injury, 10 minutes post-injury and 30 minutes post-injury images from 3 other pigs. The residual images or the reconstructed error maps show promise in detecting hemorrhages with an AUC of 0.90, 0.87, 0.62 immediately, 10 minutes post-injury and 30 minutes post-injury respectively with an overall AUC of 0.83.

A Compensation Network With Error Mapping for Robust Remote Photoplethysmography in Noise-Heavy Conditions

Remote photoplethysmography (rPPG) measurement has received much attention due to its attractive applications in contactless physiological monitoring. It calculates the blood volume pulse from color signals with a normal camera. Artifacts caused by motion and illumination changes must be accounted for, especially in real-world usage. In this article, we propose a novel framework that includes an error compensation neural network after conventional signal-based rPPG extraction to improve the measuring results. Our main idea is to leverage facial image factors and frequency spectrum features to characterize the relevance between the current measurement conditions and measurement error; given inaccurate measurements, the compensation network computes the corresponding correction term to improve performance. The implementation of this compensation network includes a feature extractor based mainly on 1-D convolution kernels and a recurrent structure to account for temporal interference. Results of experiments conducted on two open datasets and two self-constructed datasets recorded for compact car drivers and passengers with driving disturbance demonstrate significant improvements over state-of-the-art methods, especially in challenging cases. The percentage of the mean absolute error with respect to that based on conventional signal-based methods improves by up to 72.3% for drivers and 66.7% for passengers.

Collaborative boundary-aware context encoding networks for error map prediction

Accurately assessing the medical image segmentation quality of the automatically generated predictions is essential for guaranteeing the reliability of the results of computer-assisted diagnosis (CAD). Many researchers have studied segmentation quality estimation without labeled ground truths. Recently, a novel idea is proposed, which transforms segmentation quality assessment (SQA) into the pixel-wise or voxel-wise error map segmentation task. However, the simple application of vanilla segmentation structures in medical domain fails to achieve satisfactory error segmentation results. In this paper, we propose collaborative boundary-aware context encoding networks called EP-Net for error segmentation task. Specifically, we propose a collaborative feature transformation branch for better feature fusion between images and masks, and precise localization of error regions. Further, we propose a context encoding module to utilize the global predictor from the error map to enhance the feature representation and regularize the networks. Extensive experiments on IBSR V2.0 dataset, ACDC dataset and M&Ms dataset demonstrate that EP-Net achieves better error segmentation results compared with the traditional segmentation patterns. Based on error prediction results, we obtain a proxy metric of segmentation quality, which has high Pearson correlation coefficient with the real segmentation accuracy on all datasets.

3D Fourier ghost imaging via semi-calibrated photometric stereo.

We achieved three-dimensional (3D) computational ghost imaging with multiple photoresistors serving as single-pixel detectors using the semi-calibrated lighting approach. We performed imaging in the spatial frequency domain by having each photoresistor obtain the Fourier spectrum of the scene at a low spectral coverage ratio of 5%. To retrieve a depth map of a scene, we inverted, apodized, and applied semi-calibrated photometric stereo (SCPS) to the spectra. At least 93.5% accuracy was achieved for the 3D results of the apodized set of images applied with SCPS in comparison with the ground truth. Furthermore, intensity error map statistics obtained at least 97.0% accuracy for the estimated surface normals using our method. Our system does not need special calibration objects or any additional optical components to perform accurate 3D imaging, making it easily adaptable. Our method can be applied in current imaging systems where multiple detectors operating at any wavelength are used for two-dimensional (2D) imaging, such as imaging cosmological objects. Employing the idea of changing light patterns to illuminate a target scene and having stored information about these patterns, the data retrieved by one detector will give the 2D information while the multiple-detector system can be used to get a 3D profile.

The effect of pre-aggregation scale on spatially adaptive filters

Choropleth mapping continues to be a dominant mapping technique despite suffering from the Modifiable Areal Unit Problem (MAUP), which may distort disease risk patterns when different administrative units are used. Spatially adaptive filters (SAF) are one mapping technique that can address the MAUP, but the limitations and accuracy of spatially adaptive filters are not well tested. Our work examines these limitations by using varying levels of data aggregation using a case study of geocoded breast cancer screening data and a synthetic georeferenced population dataset that allows us to calculate SAFs at the individual-level. Data were grouped into four administrative boundaries (i.e., county, Zip Code Tabulated Areas, census tracts, and census blocks) and compared to individual-level data (control). Correlation assessed the similarity of SAFs, and map algebra calculated error maps compared to control. This work describes how pre-aggregation affects the level of spatial detail, map patterns, and over and under-prediction.