Candidate Patches Research Articles

Pre-Diabetic Retinopathy identification using hybridGenetic Algorithm-Neural Network classifier

Diabetic retinopathy (DR) is one of the main prevalent diabetes problems, causing blurry vision and degeneration amongst adults of working age. The first symptoms of DR include Microaneurysms (MA). A Genetic Algorithm-Artificial Neural Network (GA-NN) technique is developed for early diagnosis of DR. There are five steps of the proposed framework. Image pre-processing is achieved using r-polynomial transformation. In the extraction, the K-means algorithm is used to segment blood vessels, and candidate patches were generated. Shape attributes, GLCM and LBP features have been derived from excluded blood vessel image and from patches separately. To achieve independent classification, GA-NN classifiers is employed. The ultimate decision system projects the MA or non-MA class labels by plurality voting for eachclassifier. This methodology was tested on two databases: e-Ophtha-MA and DIARETDB1. The e-ophtha-MA and DIARETDB1 datasets had AUCs of 0.89 and 0.87, respectively, on the receiver operating characteristic (ROC) curve.

A CNN-Based High-Accuracy Registration for Remote Sensing Images

In this paper, a convolutional neural network-based registration framework is proposed for remote sensing to improve the registration accuracy between two remote-sensed images acquired from different times and viewpoints. The proposed framework consists of four stages. In the first stage, key-points are extracted from two input images—a reference and a sensed image. Then, a patch is constructed at each key-point. The second stage consists of three processes for patch matching—candidate patch pair list generation, one-to-one matched label selection, and geometric distortion compensation. One-to-one matched patch pairs between two images are found, and the exact matching is found by compensating for geometric distortions in the matched patch pairs. A global geometric affine parameter set is computed using the random sample consensus algorithm (RANSAC) algorithm in the third stage. Finally, a registered image is generated after warping the input sensed image using the affine parameter set. The proposed high-accuracy registration framework is evaluated using the KOMPSAT-3 dataset by comparing the conventional frameworks based on machine learning and deep-learning-based frameworks. The proposed framework obtains the least root mean square error value of 34.922 based on all control points and achieves a 68.4% increase in the matching accuracy compared with the conventional registration framework.

Learning Deep Patch representation for Probabilistic Graphical Model-Based Face Sketch Synthesis

Face sketch synthesis has a wide range of applications in both digital entertainment and law enforcement. State-of-the-art examplar-based methods typically exploit a Probabilistic Graphical Model (PGM) to represent the joint probability distribution over all of the patches selected from a set of training data. However, these methods suffer from two main shortcomings: (1) most of these methods capture the evidence between patches in pixel-level, which lead to inaccurate parameter estimation under bad environment conditions such as light variations and clutter backgrounds; (2) the assumption that a photo patch and its corresponding sketch patch share similar geometric manifold structure is not rigorous. It has shown that deep convolutional neural network (CNN) has outstanding performance in learning to extract high-level feature representation. Therefore, we extract uniform deep patch representations of test photo patches and training sketch patches from a specially designed CNN model to replace pixel intensity, and directly match between them, which can help select better candidate patches from training data as well as improve parameter learning process. In this way, we investigate a novel face sketch synthesis method called DPGM that combines generative PGM and discriminative deep patch representation, which can jointly model the distribution over the parameters for deep patch representation and the distribution over the parameters for sketch patch reconstruction. Then, we apply an alternating iterative optimization strategy to simultaneously optimize two kinds of parameters. Therefore, both the representation capability of deep patch representation and the reconstruction ability of sketch patches can be boosted. Eventually, high quality reconstructed sketches which is robust against light variations and clutter backgrounds can be obtained. Extensive experiments on several benchmark datasets demonstrate that our method can achieve superior performance than other state-of-the-art methods, especially under the case of bad light conditions or clutter backgrounds.

Fixing Multiple Type Errors in Model Transformations with Alternative Oracles to Test Cases.

This paper addresses the issue of correcting type errors in model transformations in realistic scenarios where neither predefined patches nor behavior-safe guards such as test suites are available. Instead of using predefined patches targeting isolated errors of specific categories, we propose to explore the space of possible patches by combining basic edit operations for model transformation programs. To guide the search, we define two families of objectives: one to limit the number of type errors and the other to preserve the transformation behavior. To approximate the latter, we study two objectives: minimizing the number of changes and keeping the changes local. Additionally, we define four heuristics to refine candidate patches to increase the likelihood of correcting type errors while preserving the transformation behavior. We implemented our approach for the ATL language using the evolutionary algorithm NSGA-II, and performed an evaluation based on three published case studies. The evaluation results show that our approach was able to automatically correct on average more than82% of type errors for two cases and more than 56% for the third case.

Evaluating the usage of fault localization in automated program repair: an empirical study

Fault localization techniques are originally proposed to assist in manual debugging by generally producing a rank list of suspicious locations. With the increasing popularity of automated program repair, the fault localization techniques have been introduced to effectively reduce the search space of automated program repair. Unlike developers who mainly focus on the rank information, current automated program repair has two strategies to use the fault localization information: suspiciousness-first algorithm (SFA) based on the suspiciousness accuracy and rank-first algorithm (RFA) relying on the rank accuracy. However, despite the fact that the two different usages are widely adopted by current automated program repair and may result in different repair results, little is known about the impacts of the two strategies on automated program repair. In this paper we empirically compare the performance of SFA and RFA in the context of automated program repair. Specifically, we implement the two strategies and six well-studied fault localization techniques into four state-of-the-art automated program repair tools, and then use these tools to perform repair experiments on 60 real-world bugs from Defects4J. Our study presents a number of interesting findings: RFA outperforms SFA in 70.02% of cases when measured by the number of candidate patches generated before a valid patch is found (NCP), while SFA performs better in parallel repair and patch diversity; the performance of SFA can be improved by increasing the suspiciousness accuracy of fault localization techniques; finally, we use SimFix that deploys SFA to successfully repair four extra Defects4J bugs which cannot be repaired by SimFix originally using RFA. These observations provide a new perspective for future research on the usage and improvement of fault localization in automated program repair.

Smart Contract Repair

Smart contracts are automated or self-enforcing contracts that can be used to exchange assets without having to place trust in third parties. Many commercial transactions use smart contracts due to their potential benefits in terms of secure peer-to-peer transactions independent of external parties. Experience shows that many commonly used smart contracts are vulnerable to serious malicious attacks, which may enable attackers to steal valuable assets of involving parties. There is, therefore, a need to apply analysis and automated repair techniques to detect and repair bugs in smart contracts before being deployed. In this work, we present the first general-purpose automated smart contract repair approach that is also gas-aware. Our repair method is search-based and searches among mutations of the buggy contract. Our method also considers the gas usage of the candidate patches by leveraging our novel notion of gas dominance relationship . We have made our smart contract repair tool SCRepair available open-source, for investigation by the wider community.

Super‐resolution image reconstruction using molecular docking

Molecular-docking is an essential tool in the drug designing process, where a small molecule ligand (drug) binds with disease-causing protein molecule to prevent its further activity. Docking process helps in predicting the most appropriate configuration and the optimal interaction energy between the interacting molecules (ligand and protein) to form a stable complex. Based on this idea, a new learning-based single image super-resolution reconstruction (LSI-SRR) method is proposed here. Estimation of a high resolution (HR) patch is achieved by optimising the interaction energy between the input low resolution patch and its corresponding candidate patches appropriately chosen from the training image dataset via Genetic algorithm. Structured-spatiogram based measure; a new and competent similarity criterion is proposed to select potentially efficient training images which encompass better statistical and structural co-relation with the input image. The proposed method is tested on synthetic and real-time images at different magnification factors. Performance analysis of the proposed work is compared with some of the representative state-of-the-art LSI-SRR methods. Experimental results demonstrate that the proposed method produces HR images with enhanced image details, minimal artefacts and most importantly enables an efficient trade-off between the image qualities to speed than the competing methods.

Data Embeddable Texture Synthesis with Fast Data Extraction

Texture synthesis which grows source texture image into a bigger texture image is applied for data embedding, instead of using a readily available digital image as a carrier for data. The patch based texture synthesis is used to embed the data in patches. The selection of a suitable patch for texture synthesis is dependent on data to be hidden. At receiver, embedded data is recovered along with source texture. The efforts are taken to improve the speed of the data extraction. In the process of data extraction, the matched candidate patch is identified at every patch location of stego synthetic texture image. If it is done by referring to the energy value of the kernel region than by comparing the whole patch with all candidate patches for similarity, data extraction is almost 40 times faster.

A Two-Stage Multiple Instance Learning Framework for the Detection of Breast Cancer in Mammograms.

Mammograms are commonly employed in the large scale screening of breast cancer which is primarily characterized by the presence of malignant masses. However, automated image-level detection of malignancy is a challenging task given the small size of the mass regions and difficulty in discriminating between malignant, benign mass and healthy dense fibro-glandular tissue. To address these issues, we explore a two-stage Multiple Instance Learning (MIL) framework. A Convolutional Neural Network (CNN) is trained in the first stage to extract local candidate patches in the mammograms that may contain either a benign or malignant mass. The second stage employs a MIL strategy for an image level benign vs. malignant classification. A global image-level feature is computed as a weighted average of patch-level features learned using a CNN. Our method performed well on the task of localization of masses with an average Precision/Recall of 0.76/0.80 and achieved an average AUC of 0.91 on the image-level classification task using a five-fold cross-validation on the INbreast dataset. Restricting the MIL only to the candidate patches extracted in Stage 1 led to a significant improvement in classification performance in comparison to a dense extraction of patches from the entire mammogram.

A New Single-Image Super-Resolution Using Efficient Feature Fusion and Patch Similarity in Non-Euclidean Space

Efficient trade-off between the reconstruction qualities and the processing time of any single-image super-resolution reconstruction (SISRR) approach is critically influenced by two major aspects. These aspects are (i) appropriate representation of image patch in feature space and (ii) effective searching of candidate patches from the pool of training patches or learned dictionary. This paper proposes a neighbor embedding-based SISRR method. Novelties of our work include integration of (i) efficient feature mapping scheme which fuses multiple correlated features naturally, (ii) faster searching of candidate patches by measuring the patch correlation in non-Euclidean space and (iii) adaptive selection of neighborhood size using patch characteristic. Correlation among features is modeled via global covariance matrix, and the fusion process enables to preserve sufficient structural, spatial correlation among patches. Distance functions based on notion of generalized eigenvalue are used for measuring patch similarity which support faster searching of candidate patches. Performance analysis of the suggested method is compared with some of the competent state-of-the-art methodologies. From the simulated result analysis, proposed work is found to be outperforming in terms of sharpened image details with diminished effect of artifacts at a reasonable computational burden.

BM3D image denoising algorithm based on an adaptive filtering

Block-matching and 3D filtering algorithm (BM3D) is the current state-of-the-art for image denoising. This algorithm has a high capacity to achieve better noise removal results as compared with other existing algorithms. Nevertheless, there is still much room for improvement in this algorithm to achieve more attractive results. To address the shortcomings of BM3D filtering, our paper algorithm makes the following contributions: Firstly, the traditional hard-thresholding of the BM3D method is substituted by an adaptive filtering technique. This technique has a high capacity to acclimate and change according to the noise intensity. More accurately, in the proposed algorithm, soft-thresholding is applied to the high-noise areas, whereas the total variation filter is applied to the light-noise areas. The self-adaptation and stability of the proposed adaptive filtering technique have enabled this technique to achieve optimal noise reduction performance and preserve the high spatial frequency detail (e.g. sharp edges). Secondly, since too small threshold leaves the most amount of the noise without removing, in contrast, a too large threshold fails to maintain the significant information of the image such as edges. Accordingly, in our proposed algorithm, applying the adaptive filtering function in the first stage is based on an adaptive threshold. This threshold is adaptable and changeable according to the amount of the noise. Thirdly, an Adaptive Weight Function (AWF) that depends on the spatial distance between the reference patch and its candidate patches, is adopted in the proposed dissimilarity measurement. When the distance between the reference patch and the candidate patch is small enough (nearby patches), AWF adopts the proposed dissimilarity measurement in computing this distance. On the other hand, when the distance between the reference patch and the candidate patch is large enough (where the candidate patches are located out of the region of the reference patch), AWF adopts the k-means clustering and the Formula (21) in computing this distance. The k-means clustering is adopted at the last estimate. Utilizing the k-means clustering to partition the image into several regions and identify the boundaries between these regions obliges the block matching to search within the region of the reference patch, which leads to reducing the risk of finding poor matching. Our proposed filter is tested on various digital images for different filtering quality measures. This filter shows significant improvements over BM3D filtering in terms of visual quality, Peak Signal-to-Noise Ratio (PSNR) index, and Structural Similarity (SSIM) index.

Microaneurysms detection in color fundus images using machine learning based on directional local contrast

BackgroundAs one of the majorcomplications of diabetes, diabetic retinopathy (DR) is a leadingcause of visual impairment and blindness due to delayed diagnosisand intervention. Microaneurysms appear as the earliest symptom ofDR. Accurate and reliable detection of microaneurysms in colorfundus images has great importance for DR screening.MethodsA microaneurysms' detection methodusing machine learning based on directional local contrast (DLC) isproposed for the early diagnosis of DR. First, blood vessels wereenhanced and segmented using improved enhancement function based onanalyzing eigenvalues of Hessian matrix. Next, with blood vesselsexcluded, microaneurysm candidate regions were obtained using shapecharacteristics and connected components analysis. After imagesegmented to patches, the features of each microaneurysm candidatepatch were extracted, and each candidate patch was classified intomicroaneurysm or non-microaneurysm. The main contributions of ourstudy are (1) making use of directional local contrast inmicroaneurysms' detection for the first time, which does make sensefor better microaneurysms' classification. (2) Applying threedifferent machine learning techniques for classification andcomparing their performance for microaneurysms' detection. Theproposed algorithm was trained and tested on e-ophtha MA database,and further tested on another independent DIARETDB1 database.Results of microaneurysms' detection on the two databases wereevaluated on lesion level and compared with existing algorithms.ResultsThe proposed method has achieved better performance compared with existing algorithms on accuracy and computation time. On e-ophtha MA and DIARETDB1 databases, the area under curve (AUC) of receiver operating characteristic (ROC) curve was 0.87 and 0.86, respectively. The free-response ROC (FROC) score on the two databases was 0.374 and 0.210, respectively. The computation time per image with resolution of 2544×1969, 1400×960 and 1500×1152 is 29 s, 3 s and 2.6 s, respectively.ConclusionsThe proposed methodusing machine learning based on directional local contrast of imagepatches can effectively detect microaneurysms in color fundus imagesand provide an effective scientific basis for early clinical DRdiagnosis.

Hydrogel-forming microneedle arrays as a therapeutic option for transdermal esketamine delivery

Treatment resistant depression is, by definition, difficult to treat using standard therapeutic interventions. Recently, esketamine has been shown as a viable rescue treatment option in patients in depressive crisis states. However, IV administration is associated with a number of drawbacks and advanced delivery platforms could provide an alternative parenteral route of esketamine dosing in patients. Hydrogel-forming microneedle arrays facilitate transdermal delivery of drugs by penetrating the outer layer of the skins surface, absorbing interstitial skin fluid and swelling. This subsequently facilitates permeation of medicines into the dermal microcirculation. This paper outlines the in vitro formulation development for hydrogel-forming microneedle arrays containing esketamine. Analytical methods for the detection and quantitation of esketamine were developed and validated according to International Conference on Harmonisation standards. Hydrogel-forming microneedle arrays were fully characterised for their mechanical strength and skin insertion properties. Furthermore, a series of esketamine containing polymeric films and lyophilised reservoirs were assessed as drug reservoir candidates. Dissolution testing and content drug recovery was carried out, followed by permeation studies using 350 μm thick neonatal porcine skin in modified Franz cell apparatus. Lead reservoir candidates were selected based on measured physicochemical properties and brought forward for testing in female Sprague-Dawley rats. Plasma samples were analysed using reverse phase high performance liquid chromatography for esketamine. Both polymeric film and lyophilised reservoirs candidate patches achieved esketamine plasma concentrations higher than the target concentration of 0.15–0.3 μg/ml over 24 h. Mean plasma concentrations in rats, 24 h post-application of microneedle patches with drug reservoir F3 and LW3, were 0.260 μg/ml and 0.498 μg/ml, respectively. This developmental study highlights the potential success of hydrogel-forming microneedle arrays as a transdermal drug delivery platform for ESK and supports moving to in vivo tests in a larger animal model.

Fast minutiae extractor using neural network

In this paper, we propose a fast and reliable neural network-based algorithm for fingerprint minutiae extraction. In particular, our algorithm involves a two-stage process: in the first stage, a network generates candidate patches in which minutiae may exist; in the second stage, another network extracts minutiae from every patch.These two networks share a common part to reduce the running time. Moreover, we analyze the properties of fingerprint images and propose a principle for designing efficient networks for minutiae extraction. For efficiency, our algorithm extracts minutiae directly from raw fingerprint images, without traditional pre-processes. Another benefit of this design is that the networks only require datasets with minutiae labels for training. On the public fingerprint datasets (FVC 2002 and 2004), our algorithm requires 26 ms on average to extract minutiae from one fingerprint on a single GPU. Compared with other neural network-based algorithms, our algorithm runs approximately 10 times faster and does not lose substantial accuracy.

Progressive learning for weakly supervised fine-grained classification

Despite fine-grained image classification has made considerable progress, it still remains a challenging task due to the difficulty of finding subtle distinctions. Most existing methods solve this problem by selecting the top-N highest scores’ discriminative patches from candidate patches at one time. However, since the classification network often highlights small and sparse regions, the selected patches with the lower rank may contain noise information.To address this problem and ensure the diversity of fine-grained features, we propose a progressive patch localization module (PPL) to find the discriminative patches more accurately. Specifically, this work employs the classification model to find first most discriminative patch, then removes the most salient region to help the localization of the next most discriminative patch, and the top-K discriminative patches can be found by repeating this procedure. In addition, in order to further improve the representational power of patch-level features, we propose a feature calibration module (FCM). This module employs the global information to selectively emphasize discriminative features and suppress useless information, which can obtain more robust and discriminative local feature representations and then help classification network achieve better performance. Extensive experiments are conducted to show the substantial improvements of our method on three benchmark datasets.

A novel patch-based nonlinear matrix completion algorithm for image analysis through convolutional neural network

Matrix completion is extensively studied due to its wide applications in science and technology. In this paper, we concentrate our study on the matrix completion problem for image analysis tasks due to their immense importance and pervasive use in many fields. A rich collection of models has been proposed to capture both linear and nonlinear relationships latent in a matrix. Even though nonlinear models possess more powerful matrix completion capabilities than their linear counterparts, these models generally carry higher model complexity and tuning difficulties. To take advantage of the superior discriminative power offered by nonlinear models while curbing its deployment overheads, this paper proposes a novel nonlinear matrix completion model utilizing a deep learning-based approach. In contrast to existing nonlinear models, the new model carefully explores and exploits spatial locality among adjacent matrix elements exhibited in patches of various sizes and locations in a target matrix. Building upon this idea, a new patch-based nonlinear matrix completion algorithm is designed. The algorithm leverages a convolutional neural network to learn the predictive relationship between a matrix element and its surrounding elements through an end-to-end trainable fashion, leading to a capable and easy-to-deploy nonlinear matrix completion solution. To identify an optimal patch size suited for tackling a given matrix completion task without exhaustively enumerating all candidate patch sizes, the new algorithm is coupled with a fast stochastic search procedure, yielding a good trade-off between computational efficiency and accuracy. Extensive experiments are conducted to validate the effectiveness and advantages of the proposed algorithm for nonlinear matrix completion problem in comparison with a series of state-of-the-art algorithms. Experimental results consistently demonstrate the superiority of the new algorithm in completing images with a variety of random and textual noises.

Automatic Lumbar Spine Tracking Based on Siamese Convolutional Network.

Deep learning has demonstrated great success in various computer vision tasks. However, tracking the lumbar spine by digitalized video fluoroscopic imaging (DVFI), which can quantitatively analyze the motion mode of the spine to diagnose lumbar instability, has not yet been well developed due to the lack of steady and robust tracking method. The aim of this work is to automatically track lumbar vertebras with rotated bounding boxes in DVFI sequences. Instead of distinguishing vertebras using annotated lumbar images or sequences, we train a full-convolutional siamese neural network offline to learn generic image features with transfer learning. The siamese network is trained to learn a similarity function that compares the labeled target from the initial frame with the candidate patches from the current frame. The similarity function returns a high score if the two images depict the same object. Once learned, the similarity function is used to track a previously unseen object without any adapting online. Our tracker is performed by evaluating the candidate rotated patches sampled around the previous target's position and presents rotated bounding boxes to locate the lumbar spine from L1 to L4. Results indicate that the proposed tracking method can track the lumbar vertebra steadily and robustly. The study demonstrates that the lumbar tracker based on siamese convolutional network can be trained successfully without annotated lumbar sequences.

An Empirical Study on Learning Bug-Fixing Patches in the Wild via Neural Machine Translation

Millions of open source projects with numerous bug fixes are available in code repositories. This proliferation of software development histories can be leveraged to learn how to fix common programming bugs. To explore such a potential, we perform an empirical study to assess the feasibility of using Neural Machine Translation techniques for learning bug-fixing patches for real defects. First, we mine millions of bug-fixes from the change histories of projects hosted on GitHub in order to extract meaningful examples of such bug-fixes. Next, we abstract the buggy and corresponding fixed code, and use them to train an Encoder-Decoder model able to translate buggy code into its fixed version. In our empirical investigation, we found that such a model is able to fix thousands of unique buggy methods in the wild. Overall, this model is capable of predicting fixed patches generated by developers in 9--50% of the cases, depending on the number of candidate patches we allow it to generate. Also, the model is able to emulate a variety of different Abstract Syntax Tree operations and generate candidate patches in a split second.

Motion-Compensated Spatio-Temporal Filtering for Multi-Image and Multimodal Super-Resolution

The classical multi-image super-resolution model assumes that the super-resolved image is related to the low-resolution frames by warping, convolution and downsampling. State-of-the-art algorithms either use explicit registration to fuse the information for each pixel in its trajectory or exploit spatial and temporal similarities. We propose to combine both ideas, making use of inter-frame motion and exploiting spatio-temporal redundancy with patch-based techniques. We introduce a non-linear filtering approach that combines patches from several frames not necessarily belonging to the same pixel trajectory. The selection of candidate patches depends on a motion-compensated 3D distance, which is robust to noise and aliasing. The selected 3D volumes are then sliced per frame, providing a collection of 2D patches which are finally averaged depending on their similarity to the reference one. This makes the upsampling strategy robust to flow inaccuracies and occlusions. Total variation and nonlocal regularization are used in the deconvolution stage. The experimental results demonstrate the state-of-the-art performance of the proposed method for the super-resolution of videos and light-field images. We also adapt our approach to multimodal sequences when some additional data at the desired resolution is available.

A two-stage method for automated detection of ring-like endosomes in fluorescent microscopy images.

Endosomes are subcellular organelles which serve as important transport compartments in eukaryotic cells. Fluorescence microscopy is a widely applied technology to study endosomes at the subcellular level. In general, a microscopy image can contain a large number of organelles and endosomes in particular. Detecting and annotating endosomes in fluorescence microscopy images is a critical part in the study of subcellular trafficking processes. Such annotation is usually performed by human inspection, which is time-consuming and prone to inaccuracy if carried out by inexperienced analysts. This paper proposes a two-stage method for automated detection of ring-like endosomes. The method consists of a localization stage cascaded by an identification stage. Given a test microscopy image, the localization stage generates a voting-map by locally comparing the query endosome patches and the test image based on a bag-of-words model. Using the voting-map, a number of candidate patches of endosomes are determined. Subsequently, in the identification stage, a support vector machine (SVM) is trained using the endosome patches and the background pattern patches. Each of the candidate patches is classified by the SVM to rule out those patches of endosome-like background patterns. The performance of the proposed method is evaluated with real microscopy images of human myeloid endothelial cells. It is shown that the proposed method significantly outperforms several state-of-the-art competing methods using multiple performance metrics.