Fast Rotation-Free Feature-Based Image Registration Using Improved N-SIFT and GMM-Based Parallel Optimization

Robust endoscopic image mosaicking via fusion of multimodal estimation

We propose an endoscopic image mosaicking algorithm that is robust to light conditioning changes, specular reflections, and feature-less scenes. These conditions are especially common in minimally invasive surgery where the light source moves with the camera to dynamically illuminate close range scenes. This makes it difficult for a single image registration method to robustly track camera motion and then generate consistent mosaics of the expanded surgical scene across different and heterogeneous environments. Instead of relying on one specialised feature extractor or image registration method, we propose to fuse different image registration algorithms according to their uncertainties, formulating the problem as affine pose graph optimisation. This allows to combine landmarks, dense intensity registration, and learning-based approaches in a single framework. To demonstrate our application we consider deep learning-based optical flow, hand-crafted features, and intensity-based registration, however, the framework is general and could take as input other sources of motion estimation, including other sensor modalities. We validate the performance of our approach on three datasets with very different characteristics to highlighting its generalisability, demonstrating the advantages of our proposed fusion framework. While each individual registration algorithm eventually fails drastically on certain surgical scenes, the fusion approach flexibly determines which algorithms to use and in which proportion to more robustly obtain consistent mosaics.

An Automatic Detection Method for Cutting Path of Chips in Wafer

Microscopic imaging is easily affected by the strength of illumination, and the chip surface qualities of different wafers are different. Therefore, wafer images have defects such as uneven brightness distribution, obvious differences in chip region characteristics, etc., which affect the positioning accuracy of the wafer cutting path. For this reason, this thesis proposes an automatic chip-cutting path-planning method in the wafer image of the Glass Passivation Parts (GPPs) process without a mark. First, the wafer image is calibrated for brightness. Then, the template matching algorithm is used to determine the chip region and the center of gravity position of the chip region. We find the position of the geometric feature (interlayer) in the chip region, and the interlayer is used as an auxiliary location to determine the final cutting path. The experiment shows that the image quality can be improved, and chip region features can be highlighted when preprocessing the image with brightness calibration. The results show that the average deviation of the gravity coordinates of the chip region in the x direction is 2.82 pixels. We proceeded by finding the interlayer in the chip region, marking it with discrete points, and using the improved Random Sample Consensus (RANSAC) algorithm to remove the abnormal discrete points and fit the remaining discrete points. The average fitting error is 0.8 pixels, which is better than the least squares method (LSM). The cutting path location algorithm proposed in this paper can adapt to environmental brightness changes and different qualities of chips, accurately and quickly determine the cutting path, and improve the chip cutting yield.

Lung Respiratory Motion Estimation Based on Fast Kalman Filtering and 4D CT Image Registration

Respiratory motion estimation is an important part in image-guided radiation therapy and clinical diagnosis. However, most of the respiratory motion estimation methods rely on indirect measurements of external breathing indicators, which will not only introduce great estimation errors, but also bring invasive injury for patients. In this paper, we propose a method of lung respiratory motion estimation based on fast Kalman filtering and 4D CT image registration (LRME-4DCT). In order to perform dynamic motion estimation for continuous phases, a motion estimation model is constructed by combining two kinds of GPU-accelerated 4D CT image registration methods with fast Kalman filtering method. To address the high computational requirements of 4D CT image sequences, a multi-level processing strategy is adopted in the 4D CT image registration methods, and respiratory motion states are predicted from three independent directions. In the DIR-lab dataset and POPI dataset with 4D CT images, the average target registration error (TRE) of the LRME-4DCT method can reach 0.91 mm and 0.85 mm respectively. Compared with traditional estimation methods based on pair-wise image registration, the proposed LRME-4DCT method can estimate the physiological respiratory motion more accurately and quickly. Our proposed LRME-4DCT method fully meets the practical clinical requirements for rapid dynamic estimation of lung respiratory motion.

Different Medical Image Registration Techniques: A Comparative Analysis

Background:

Image registration provides major role in real world applications and classic digital image processing. Image registration is carried out for more than one image and this image was captured from a different location, different sensors, different time and different viewpoints.

Discussion:

This paper deals with the comparative analysis of various registration techniques and here six registration techniques depending upon intensity, phase correlation, image feature, area, control points and mutual information are compared. Comparative analysis for different methodologies shows the advantages of one method over the other methods. The foremost objective of this paper is to deliver a complete reference source for the scholars interested in registration, irrespective of specific application extents.

Conclusion:

Finally performance analyses are evaluated for the medical datasets and comparison is graphically shown with the MATLAB simulation tool.

Dual Model Medical Invoices Recognition

Hospitals need to invest a lot of manpower to manually input the contents of medical invoices (nearly 300,000,000 medical invoices a year) into the medical system. In order to help the hospital save money and stabilize work efficiency, this paper designed a system to complete the complicated work using a Gaussian blur and smoothing–convolutional neural network combined with a recurrent neural network (GBS-CR) method. Gaussian blur and smoothing (GBS) is a novel preprocessing method that can fix the breakpoint font in medical invoices. The combination of convolutional neural network (CNN) and recurrent neural network (RNN) was used to raise the recognition rate of the breakpoint font in medical invoices. RNN was designed to be the semantic revision module. In the aspect of image preprocessing, Gaussian blur and smoothing were used to fix the breakpoint font. In the period of making the self-built dataset, a certain proportion of the breakpoint font (the font of breakpoint is 3, the original font is 7) was added, in this paper, so as to optimize the Alexnet–Adam–CNN (AA-CNN) model, which is more suitable for the recognition of the breakpoint font than the traditional CNN model. In terms of the identification methods, we not only adopted the optimized AA-CNN for identification, but also combined RNN to carry out the semantic revisions of the identified results of CNN, meanwhile further improving the recognition rate of the medical invoices. The experimental results show that compared with the state-of-art invoice recognition method, the method presented in this paper has an average increase of 10 to 15 percentage points in recognition rate.

Automated Neuron Detection in High-Content Fluorescence Microscopy Images Using Machine Learning

The study of neuronal morphology in relation to function, and the development of effective medicines to positively impact this relationship in patients suffering from neurodegenerative diseases, increasingly involves image-based high-content screening and analysis. The first critical step toward fully automated high-content image analyses in such studies is to detect all neuronal cells and distinguish them from possible non-neuronal cells or artifacts in the images. Here we investigate the performance of well-established machine learning techniques for this purpose. These include support vector machines, random forests, k-nearest neighbors, and generalized linear model classifiers, operating on an extensive set of image features extracted using the compound hierarchy of algorithms representing morphology, and the scale-invariant feature transform. We present experiments on a dataset of rat hippocampal neurons from our own studies to find the most suitable classifier(s) and subset(s) of features in the common practical setting where there is very limited annotated data for training. The results indicate that a random forests classifier using the right feature subset ranks best for the considered task, although its performance is not statistically significantly better than some support vector machine based classification models.

A Review of Point Set Registration: From Pairwise Registration to Groupwise Registration

This paper presents a comprehensive literature review on point set registration. The state-of-the-art modeling methods and algorithms for point set registration are discussed and summarized. Special attention is paid to methods for pairwise registration and groupwise registration. Some of the most prominent representative methods are selected to conduct qualitative and quantitative experiments. From the experiments we have conducted on 2D and 3D data, CPD-GL pairwise registration algorithm and JRMPC groupwise registration algorithm seem to outperform their rivals both in accuracy and computational complexity. Furthermore, future research directions and avenues in the area are identified.

Face Recognition Using the SR-CNN Model

In order to solve the problem of face recognition in complex environments being vulnerable to illumination change, object rotation, occlusion, and so on, which leads to the imprecision of target position, a face recognition algorithm with multi-feature fusion is proposed. This study presents a new robust face-matching method named SR-CNN, combining the rotation-invariant texture feature (RITF) vector, the scale-invariant feature transform (SIFT) vector, and the convolution neural network (CNN). Furthermore, a graphics processing unit (GPU) is used to parallelize the model for an optimal computational performance. The Labeled Faces in the Wild (LFW) database and self-collection face database were selected for experiments. It turns out that the true positive rate is improved by 10.97–13.24% and the acceleration ratio (the ratio between central processing unit (CPU) operation time and GPU time) is 5–6 times for the LFW face database. For the self-collection, the true positive rate increased by 12.65–15.31%, and the acceleration ratio improved by a factor of 6–7.

A mobile computer aided system for optic nerve head detection

BACKGROUND AND OBJECTIVE

The detection of optic nerve head (ONH) in retinal fundus images plays a key role in identifying Diabetic Retinopathy (DR) as well as other abnormal conditions in eye examinations. This paper presents a method and its associated software towards the development of an Android smartphone app based on a previously developed ONH detection algorithm. The development of this app and the use of the d-Eye lens which can be snapped onto a smartphone provide a mobile and cost-effective computer-aided diagnosis (CAD) system in ophthalmology. In particular, this CAD system would allow eye examination to be conducted in remote locations with limited access to clinical facilities.

METHODS

A pre-processing step is first carried out to enable the ONH detection on the smartphone platform. Then, the optimization steps taken to run the algorithm in a computationally and memory efficient manner on the smartphone platform is discussed.

RESULTS

The smartphone code of the ONH detection algorithm was applied to the STARE and DRIVE databases resulting in about 96% and 100% detection rates, respectively, with an average execution time of about 2 s and 1.3 s. In addition, two other databases captured by the d-Eye and iExaminer snap-on lenses for smartphones were considered resulting in about 93% and 91% detection rates, respectively, with an average execution time of about 2.7 s and 2.2 s, respectively.

Three-dimensional digital reconstruction of skin epidermis and dermis

This study describes how three-dimensional (3D) human skin tissue is reconstructed, and provides digital anatomical data for the physiological structure of human skin tissue based on large-scale thin serial sections. Human skin samples embedded in paraffin were cut serially into thin sections and then stained with hematoxylin-eosin. Images of serial sections obtained from lighting microscopy were scanned and aligned by the scale-invariant feature transform algorithm. 3D reconstruction of the skin tissue was generated using Mimics software. Fibre content, porosity, average pore diameter and specific surface area of dermis were analysed using the ImageJ analysis system. The root mean square error and mutual information based on the scale-invariant feature transform algorithm registration were significantly greater than those based on the manual registration. Fibre distribution gradually decreased from top to bottom; while porosity showed an opposite trend with irregular average pore diameter distribution. A specific surface area of the dermis showed a 'V' shape trend. Our data suggested that 3D reconstruction of human skin tissue based on large-scale serial sections could be a valuable tool for providing a highly accurate histological structure for analysis of skin tissue. Moreover, this technology could be utilized to produce tissue-engineered skin via a 3D bioprinter in the future.