Machine learning analysis of wing venation patterns accurately identifies Sarcophagidae, Calliphoridae and Muscidae fly species.

In medical, veterinary and forensic entomology, the ease and affordability of image data acquisition have resulted in whole-image analysis becoming an invaluable approach for species identification. Krawtchouk moment invariants are a classical mathematical transformation that can extract local features from an image, thus allowing subtle species-specific biological variations to be accentuated for subsequent analyses. We extracted Krawtchouk moment invariant features from binarised wing images of 759 male fly specimens from the Calliphoridae, Sarcophagidae and Muscidae families (13 species and a species variant). Subsequently, we trained the Generalized, Unbiased, Interaction Detection and Estimation random forests classifier using linear discriminants derived from these features and inferred the species identity of specimens from the test samples. Fivefold cross-validation results show a 98.56 ± 0.38% (standard error) mean identification accuracy at the family level and a 91.04 ± 1.33% mean identification accuracy at the species level. The mean F1-score of 0.89 ± 0.02 reflects good balance of precision and recall properties of the model. The present study consolidates findings from previous small pilot studies of the usefulness of wing venation patterns for inferring species identities. Thus, the stage is set for the development of a mature data analytic ecosystem for routine computer image-based identification of fly species that are of medical, veterinary and forensic importance.

Compliance Prediction for Structural Topology Optimization on the Basis of Moment Invariants and a Generalized Regression Neural Network.

Topology optimization techniques are essential for manufacturing industries, such as designing fiber-reinforced polymer composites (FRPCs) and structures with outstanding strength-to-weight ratios and light weights. In the SIMP approach, artificial intelligence algorithms are commonly utilized to enhance traditional FEM-based compliance minimization procedures. Based on an effective generalized regression neural network (GRNN), a new deep learning algorithm of compliance prediction for structural topology optimization is proposed. The algorithm learns the structural information using a fourth-order moment invariant analysis of the structural topology obtained from FEA at different iterations of classical topology optimization. A cantilever and a simply supported beam problem are used as ground-truth datasets, and the moment invariants are used as independent variables for input features. By comparing it with the well-known convolutional neural network (CNN) and deep neural network (DNN) models, the proposed GRNN model achieves a high prediction accuracy (R2 > 0.97) and drastically shortens the training and prediction cost. Furthermore, the GRNN algorithm exhibits excellent generalization ability on the prediction performance of the optimized topology with rotations and varied material volume fractions. This algorithm is promising for the replacement of the FEA calculation in the SIMP method, and can be applied to real-time optimization for advanced FRPC structure design.

Amphetamine-type stimulants (ATS) drug classification using shallow one-dimensional convolutional neural network.

Amphetamine-type stimulants (ATS) drug analysis and identification are challenging and critical nowadays with the emergence production of new synthetic ATS drugs with sophisticated design compounds. In the present study, we proposed a one-dimensional convolutional neural network (1DCNN) model to perform ATS drug classification as an alternative method. We investigate as well as explore the classification behavior of 1DCNN with the utilization of the existing novel 3D molecular descriptors as ATS drugs representation to become the model input. The proposed 1DCNN model is composed of one convolutional layer to reduce the model complexity. Besides, pooling operation that is a standard part of traditional CNN is not applied in this architecture to have more features in the classification phase. The dropout regularization technique is employed to improve model generalization. Experiments were conducted to find the optimal values for three dominant hyper-parameters of the 1DCNN model which are the filter size, transfer function, and batch size. Our findings found that kernel size 11, exponential linear unit (ELU) transfer function and batch size 32 are optimal for the 1DCNN model. A comparison with several machine learning classifiers has shown that our proposed 1DCNN has achieved comparable performance with the Random Forest classifier and competitive performance with the others.

FGFF Descriptor and Modified Hu Moment-Based Hand Gesture Recognition.

Gesture recognition has been studied for decades and still remains an open problem. One important reason is that the features representing those gestures are not sufficient, which may lead to poor performance and weak robustness. Therefore, this work aims at a comprehensive and discriminative feature for hand gesture recognition. Here, a distinctive Fingertip Gradient orientation with Finger Fourier (FGFF) descriptor and modified Hu moments are suggested on the platform of a Kinect sensor. Firstly, two algorithms are designed to extract the fingertip-emphasized features, including palm center, fingertips, and their gradient orientations, followed by the finger-emphasized Fourier descriptor to construct the FGFF descriptors. Then, the modified Hu moment invariants with much lower exponents are discussed to encode contour-emphasized structure in the hand region. Finally, a weighted AdaBoost classifier is built based on finger-earth mover's distance and SVM models to realize the hand gesture recognition. Extensive experiments on a ten-gesture dataset were carried out and compared the proposed algorithm with three benchmark methods to validate its performance. Encouraging results were obtained considering recognition accuracy and efficiency.

Stratification of risk of atherosclerotic plaque using Hu's moment invariants of segmented ultrasonic images.

Myocardial infarction is one of the major life-threatening diseases. The cause is atherosclerosis i.e. the occlusion of the coronary artery by deposition of plaque on its walls. The severity of plaque deposition in the artery depends on the characteristics of the plaque. Hence, the classification of the type of plaque is crucial for assessing the risk of atherosclerosis and predicting the chances of myocardial infarction. This paper proposes prediction of atherosclerotic risk by non-invasive ultrasound image segmentation and textural feature extraction. The intima-media complex is segmented using a snakes-based segmentation algorithm on the arterial wall in the ultrasound images. Then, the plaque is extracted from the segmented intima-media complex. The features of the plaque are obtained by computing Hu's moment invariants. Visual pattern recognition independent of position, size, orientation and parallel projection could be done using these moment invariants. For the classification of the features of the plaque, an SVM classifier is used. The performance shows improvement in accuracy using lesser number of features than previous works. The reduction in feature size is achieved by incorporating segmentation in the pre-processing stage. Tenfold cross-validation protocol is used for training and testing the classifier. An accuracy of 97.9% is obtained with only two features. This proposed technique could work as an adjunct tool in quick decision-making for cardiologists and radiologists. The segmentation step introduced in the preprocessing stage improved the feature extraction technique. An improvement in performance is achieved with much less number of features.

A deep neural framework for real-time vehicular accident detection based on motion temporal templates.

Vehicular accident prediction and detection has recently garnered curiosity and large amounts of attention in machine learning applications and related areas, due to its peculiar and fascinating application potentials in the development of Intelligent Transportation Systems (ITS) that play a pivotal role in the success of emerging smart cities. In this paper, we present a new vision-based framework for real-time vehicular accident prediction and detection, based on motion temporal templates and fuzzy time-slicing. The presented framework proceeds in a stepwise fashion, starting with automatically detecting moving objects (i.e., on-road vehicles or roadside pedestrians), followed by dynamically keep tracking of the detected moving objects based on temporal templates, clustering and supervised learning. Then, an extensive set of local features is extracted from the temporal templates of moving objects. Finally, an effective deep neural network (DNN) model is trained on the extracted features to detect abnormal vehicle behavioral patterns and thus predict an accident just before it occurs. The experiments on real-world vehicular accident videos demonstrate that the framework can yield mostly promising results by achieving a hit rate of 98.5% with a false alarm rate of 4.2% that compare very favorably to those from existing approaches, while still being able to deliver delay guarantees for realtime traffic monitoring and surveillance applications.

Robust 2D and 3D images zero - watermarking using dual Hahn moment invariants and Sine Cosine Algorithm.

In this paper, we initially provide significant improvements on the computational aspects of dual Hahn moment invariants (DHMIs) in both 2D and 3D domains. These improvements ensure the numerical stability of DHMIs for large-size images. Then, we propose an efficient method for optimizing the local parameters of dual Hahn polynomials (DHPs) when computing DHMIs using the Sine-Cosine Algorithm (SCA). DHMIs optimized via SCA are used to propose new and robust zero-watermarking scheme applied to both 2D and 3D images. On one hand, the simulation results confirm the efficiency of the proposed computation of 2D and 3D DHMIs regarding the numerical stability and invariability. Indeed, the proposed computation method of 2D DHMIs allows to reach a relative error (RE) of the order ≈10-10 for images of size 1024 × 1024 with an execution time improvement ratio exceeds 70% ( ETIR ≥ 70%), which validates the efficiently of the proposed computation method. On the other hand, the simulation and comparison outcomes clearly demonstrate the robustness of the proposed zero-watermarking scheme against various geometric attacks (translation, rotation, scaling and combined transformations), as well as against other common 2D and 3D image processing attacks (compression, filtering, noise addition, etc.).

Fast and stable computation of higher-order Hahn polynomials and Hahn moment invariants for signal and image analysis.

This article presents, on the one hand, new algorithms for the fast and stable computation of discrete orthogonal Hahn polynomials of high order (HPs) based on the elimination of all gamma and factorial functions that cause the numerical fluctuations of HPs, and based on the use of appropriate stability conditions. On the other hand, a new method for the fast and numerically stable computation of Hahn moment invariants (HMIs) is also proposed. This method is mainly based on the use of new recursive relations of HPs and of matrix multiplications when calculating HMIs. To validate the efficiency of the algorithms proposed for the calculation of HPs, several signals and large images (≥4000 × 4000) are reconstructed by Hahn moments (HMs) up to the last order with a reconstruction error tending towards zero (MSE ≃ 10-10). The efficiency of the proposed method for calculating HMIs is demonstrated on large medical images (2048 × 2048) with a very low relative error (RE ≃ 10-10). Finally, comparisons with some recent work in the literature are provided.

Component-based face recognition using statistical pattern matching analysis.

The aim of this research is to develop a fusion concept to component-based face recognition algorithms for features analysis of binary facial components (BFCs), which are invariant to illumination, expression, pose variations and partial occlusion. To analyze the features, using statistical pattern matching concepts, which are the combination of Chi-square (CSQ), Hu moment invariants (HuMIs), absolute difference probability of white pixels (AbsDifPWPs) and geometric distance values (GDVs) have been proposed for face recognition. The individual grayscale face image is cropped by applying the Viola-Jones face detection algorithm from a face database having variations in illumination, appearance, pose and partial occlusion with complex backgrounds. Doing illumination correction through histogram linearization technique, the grayscale face components such as eyes, nose and mouth regions are extracted using the 2D geometric positions. The binary face image is created by applying cumulative probability distribution function with Otsu adaptive thresholding method and then extracted BFCs such as eyes, nose and mouth regions. Five statistical pattern matching tools such as the standard deviation of CSQ values with probability of white pixels (PWPs), standard deviation of HuMIs with Hu's seven moment invariants, AbsDifPWPs and GDVs are developed for recognition purpose. GDVs are determined between two similar facial corner points (FCPs) and nine FCPs are extracted from binary whole face and BFCs. Pixel Intensity Values (PIVs) which are determined using L2 norms from grayscale values of the whole face and grayscale values of the face components. Experiment is performed using BioID Face Database on the basis of these pattern matching tools and appropriate threshold values with logical and conditional operators and gives the best expected results from true positive rate perspective.

Orthogonal moments for determining correspondence between vessel bifurcations for retinal image registration.

Retinal image registration is a necessary step in diagnosis and monitoring of Diabetes Retinopathy (DR), which is one of the leading causes of blindness. Long term diabetes affects the retinal blood vessels and capillaries eventually causing blindness. This progressive damage to retina and subsequent blindness can be prevented by periodic retinal screening. The extent of damage caused by DR can be assessed by comparing retinal images captured during periodic retinal screenings. During image acquisition at the time of periodic screenings translation, rotation and scale (TRS) are introduced in the retinal images. Therefore retinal image registration is an essential step in automated system for screening, diagnosis, treatment and evaluation of DR. This paper presents an algorithm for registration of retinal images using orthogonal moment invariants as features for determining the correspondence between the dominant points (vessel bifurcations) in the reference and test retinal images. As orthogonal moments are invariant to TRS; moment invariants features around a vessel bifurcation are unaltered due to TRS and can be used to determine the correspondence between reference and test retinal images. The vessel bifurcation points are located in segmented, thinned (mono pixel vessel width) retinal images and labeled in corresponding grayscale retinal images. The correspondence between vessel bifurcations in reference and test retinal image is established based on moment invariants features. Further the TRS in test retinal image with respect to reference retinal image is estimated using similarity transformation. The test retinal image is aligned with reference retinal image using the estimated registration parameters. The accuracy of registration is evaluated in terms of mean error and standard deviation of the labeled vessel bifurcation points in the aligned images. The experimentation is carried out on DRIVE database, STARE database, VARIA database and database provided by local government hospital in Pune, India. The experimental results exhibit effectiveness of the proposed algorithm for registration of retinal images.

Region and boundary feature estimation on ultrasound images using moment invariants.

In ultrasound images, tissues are characterized by their speckle texture. Moment-based techniques have proven their ability to capture texture features. However, in ultrasound images, the speckle size increases with the distance from the probe and in some cases the speckle has a concentric texture arrangement. We propose to use moment invariants with respect to image scale and rotation to capture the texture in such cases. Results on synthetic data show that moment invariants are able to characterize the texture but also that some moment orders are sensitive to regions and, moreover, some are sensitive to the boundaries between two different textures. This behavior seems to be very interesting to be used within some segmentation scheme dealing with a combination of regional and boundary information. In this paper we will try to prove the usability of this complementary information in a min-cut/max-flow graph cut scheme.