Neuro-pattern classifiction using zernike moments and its reduced set of features

Explicit invariance of Cartesian Zernike moments

Two important factors that determine the efficiency and reliability of a rice sorting machine are the overall processing speed and the classification accuracy. In this paper, an efficient rice sorting process which uses a subset of Zernike moments (ZM) and a multilayer perceptron is presented. Since the falling rice grains during sorting process can be in any orientation, a rotational invariant feature set is crucial in this application. Hence, the set of ZM with its inherent rotational invariance property is chosen in this context. Nevertheless, one of the main drawbacks of ZM in real-time application is its long computation time. To overcome this, a subset of ZM is selected from its original full set of 12 orders, using the combination of fuzzy ARTMAP and genetic algorithm. To further reduce the computation time, the combination of q-recursive method and Zernike polynomials’ inherent symmetry property is utilized. Hence, the processing time of the subset of ZM is significantly reduced by almost 67% while maintaining the classification accuracy as compared to computing the original full set of ZMs.

Plants play a crucial role in foodstuff, medicine, industry, and environmental protection. The skill of recognising plants is very important in some applications, including conservation of endangered species and rehabilitation of lands after mining activities. However, it is a difficult task to identify plant species because it requires specialized knowledge. Developing an automated classification system for plant species is necessary and valuable since it can help specialists as well as the public in identifying plant species easily. Shape descriptors were applied on the myDAUN dataset that contains 45 tropical shrub species collected from the University of Malaya (UM), Malaysia. Based on literature review, this is the first study in the development of tropical shrub species image dataset and classification using a hybrid of leaf shape and machine learning approach. Four types of shape descriptors were used in this study namely morphological shape descriptors (MSD), Histogram of Oriented Gradients (HOG), Hu invariant moments (Hu) and Zernike moments (ZM). Single descriptor, as well as the combination of hybrid descriptors were tested and compared. The tropical shrub species are classified using six different classifiers, which are artificial neural network (ANN), random forest (RF), support vector machine (SVM), k-nearest neighbour (k-NN), linear discriminant analysis (LDA) and directed acyclic graph multiclass least squares twin support vector machine (DAG MLSTSVM). In addition, three types of feature selection methods were tested in the myDAUN dataset, Relief, Correlation-based feature selection (CFS) and Pearson’s coefficient correlation (PCC). The well-known Flavia dataset and Swedish Leaf dataset were used as the validation dataset on the proposed methods. The results showed that the hybrid of all descriptors of ANN outperformed the other classifiers with an average classification accuracy of 98.23% for the myDAUN dataset, 95.25% for the Flavia dataset and 99.89% for the Swedish Leaf dataset. In addition, the Relief feature selection method achieved the highest classification accuracy of 98.13% after 80 (or 60%) of the original features were reduced, from 133 to 53 descriptors in the myDAUN dataset with the reduction in computational time. Subsequently, the hybridisation of four descriptors gave the best results compared to others. It is proven that the combination MSD and HOG were good enough for tropical shrubs species classification. Hu and ZM descriptors also improved the accuracy in tropical shrubs species classification in terms of invariant to translation, rotation and scale. ANN outperformed the others for tropical shrub species classification in this study. Feature selection methods can be used in the classification of tropical shrub species, as the comparable results could be obtained with the reduced descriptors and reduced in computational time and cost.

Generally, desired training set is used to make a neural network learn nonlinear relations properly. The training set consists of multiple pairs of an input vector and output one. Each input vector is given to the input layer for forward calculation, and the corresponding output vector is compared with the vector yielded from the output layer. Weights between neurons are updated based on the errors using a back propagation algorithm in backward calculation. The time required for the learning process of a neural network depends on the number of total weights and that of the input-output pairs in the training set. In the proposed learning process, after the learning is progressed, e.g., 1000 iterations, input-output pairs having had worse errors are extracted from the original training set and form a new temporary set. From the next iteration, the temporary set is applied instead of the original set. This means that only pairs with worse errors are used for updating the weights until the mean value of errors decreases to a level. After the learning using the temporary set, the original set is applied again instead of the temporary set. By alternately applying the two types of sets for learning that the calculation load for convergence can be efficiently reduced. The effectiveness of the proposed method is proved by applying to an inverse kinematics problem of an industrial robot. Generally, desired training set is used to make a neural network learn nonlinear relations properly. The training set consists of multiple pairs of an input vector and output one. Each input vector is given to the input layer for forward calculation, and the corresponding output vector is compared with the vector yielded from the output layer. Weights between neurons are updated based on the errors using a back propagation algorithm in backward calculation. The time required for the learning process of a neural network depends on the number of total weights and that of the input-output pairs in the training set. In the proposed learning process, after the learning is progressed, e.g., 1000 iterations, input-output pairs having had worse errors are extracted from the original training set and form a new temporary set. From the next iteration, the temporary set is applied instead of the original set. This means that only pairs with worse errors are used for updating the weights until the mean value of errors decreases to a level. After the learning using the temporary set, the original set is applied again instead of the temporary set. By alternately applying the two types of sets for learning that the calculation load for convergence can be efficiently reduced. The effectiveness of the proposed method is proved by applying to an inverse kinematics problem of an industrial robot.

It is shown that, using high-spatial-resolution data, very high cloud classification accuracies can be obtained with a neural network approach. A texture-based neural network classifier using only single-channel visible Landsat MSS imagery achieves an overall cloud identification accuracy of 93%. Cirrus can be distinguished from boundary layer cloudiness with an accuracy of 96%, without the use of an infrared channel. Stratocumulus is retrieved with an accuracy of 92%, cumulus at 90%. The use of the neural network does not improve cirrus classification accuracy. Rather, its main effect is in the improved separation between stratocumulus and cumulus cloudiness. The present study is based on a nonlinear, nonparametric four-layer neural network approach. A three-layer neural network architecture, the nonparametric K-nearest neighbor approach, and the linear stepwise discriminant analysis procedure are compared.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>

Aims/ objectives: To demontrate effectiveness of Zernike Moments for Image Classification. Zernike moment(ZM) is an excellent region-based moment which has attracted the attentions of many image processing researchers since its first application to image analysis. Many papers have been published on several works done on ZM but no single paper ever give a detailed information of how the computation of ZM is done from the time the image is captured to the computation of ZM. This work showed how to effectively apply ZM on RGB images. We have demonstrated the effectiveness of Zernike moment in image classification system. A neuro-genetic intelligent system has been built with PNN classifier. The feature extracted viz ZM and Geometric features were further subjected to GA to bring the best combinatorial features for optimal accuracy. The algebraic structure of our novel fitness function enabled the GA to select the best results. The 10-fold CV used enabled the whole system to be unbiased giving a classification accuracy of 90.05%. A demonstration of affine properties of ZM are comprehensively stated and explained. In summary, the ZM enabled the classifier to have improved accuracy of 91% as compared with Geometric features with 89% accuracy.

Ladar range images have attracted considerable attention in automatic target recognition fields. In this paper, Zernike moments (ZMs) are applied to classify the target of the range image from an arbitrary azimuth angle. However, ZMs suffer from high computational costs. To improve the performance of target recognition based on small samples, even-order ZMs with serial-parallel backpropagation neural networks (BPNNs) are applied to recognize the target of the range image. It is found that the rotation invariance and classified performance of the even-order ZMs are both better than for odd-order moments and for moments compressed by principal component analysis. The experimental results demonstrate that combining the even-order ZMs with serial-parallel BPNNs can significantly improve the recognition rate for small samples.

Face localization using neural network is presented in this communication. Neural network was trained with two different kinds of feature parameters vectors; Zernike moments and eigenfaces. In each case, coordinate vectors of pixels surrounding faces in images were used as target vectors on the supervised training procedure. Thus, trained neural network provides on its output layer a coordinate's vector (rho,thetas) representing pixels surrounding the face contained in treated image. This way to proceed gives accurate faces contours which are well adapted to their shapes. Performances obtained for the two kinds of training feature parameters were recorded using a quantitative measurement criterion according to experiments carried out on the XM2VTS database.

In the paper, Neural Network (NN) models for gas turbine diagnostics are studied and developed. The analyses carried out are aimed at the selection of the most appropriate NN structure for gas turbine diagnostics, in terms of computational time of the NN training phase, accuracy and robustness with respect to measurement uncertainty. In particular, feed-forward NNs with a single hidden layer trained by using a back-propagation learning algorithm are considered and tested. Moreover, Multi-Input/Multi-Output NN architectures (i.e. NNs calculating all the system outputs) are compared to Multi-Input/Single-Output NNs, each of them calculating a single output of the system. The results obtained show that NNs are robust with respect to measurement uncertainty, if a sufficient number of training patterns are used. Moreover, Multi-Input/Multi-Output NNs trained with data corrupted with measurement errors seem to be the best compromise between the computational time required for NN training phase and the NN accuracy in performing gas turbine diagnostics.

A new gait recognition method based on Zernike moments and BP neural network is proposed. Zernike moments are calculated to extract gait features based on the introduced concept of normalized gait cycle. All gait Zernike moments compose the gait feature space. PCA algorithm is used to compress Zernike moments and a new lower dimension feature space containing gait spatio-temporal features is generated. Each normalized gait cycle's Zernike moments are mapped to this new feature space and compose an eigen-matrix, whose row square error vectors are used as the gait recognition eigenvectors. BP neural network is used to classify the gait features. To increase recognition accuracy, multiple training samples and multiple inputs are used for each to be recognized gait class. Experimental results show that the method can obtain accurate gait recognition in relatively simple scenes.

The advances in biomedical imaging equipment have produced a massive amount of medical images that are generated by the different modalities. Consequently, a huge volume of data has been produced and caused a complex and time-consuming retrieving process of the relevant cases. To resolve this issue, the Content-Based Biomedical Image Retrieval (CBMIR) system is applied to retrieve the related images from the earlier patients’ databases. However, the previous handcrafted features methods that applied the CBMIR model have shown poor performance in many multimodal databases. In this paper, we focus on designing CBMIR technique using Deep Learning (DL) models. We present a new Multimodal Biomedical Image Retrieval and Classification (M-BMIRC) technique for retrieving and classifying the biomedical images from huge databases. The proposed M-BMIRC model involves three dissimilar processes as following: feature extraction, similarity measurement, and classification. It uses an ensemble of handcrafted features from Zernike Moments (ZM) and deep features from Deep Convolutional Neural Networks (DCNN) for feature extraction process. Additionally, the Hausdorff Distance based similarity measure is employed to identify the resemblance between the queried image and the images that exist in the database. Moreover, the classification process gets executed on the retrieval images using the Probabilistic Neural Network (PNN) model, which allocates the class labels of the tested images. Finally, the experimental studies are conducted using two benchmark medical datasets and the results ensure the superior performance of the proposed model in terms of different measures include Average Precision Rate (APR), Average Recall Rate (ARR), F-score, accuracy, and Computation Time (CT).

With the development of 3D printing technology and lots of available 3D model digital files in the form of stl, 3ds, obj, etc., on the internet, there are many illegally distributed 3D model digital files which are protected by copyright law, and some are already printed. In this paper, we propose a method that matching shape between printed 3D model and digital 3D model based on 2D views and Zernike Moments. The printed 3D model is first photographed into several pictures by a camera, the digital 3D model is also captured into several rendering images by a virtual camera using a computer program. Next, both sides of the images are converted into binary images with denoising image processing. Then, Zernike Moments features are extracted from the binary images. Finally, we can yield similarity between the printed one and digital one by comparing the Zernike Moments.

Face recognition is an interesting field of computer vision with many commercial and ?scientific applications. It is considered as a very hot topic and challenging problem at the ?moment. Many methods and techniques have been proposed and applied for this purpose, ?such as neural networks, PCA, Gabor filtering, etc. Each approach has its weaknesses as ?well as its points of strength. This paper introduces a highly efficient method for the ?recognition of human faces in digital images using a new feature extraction method that ?combines the global and local information in different views (poses) of facial images. ?Feature extraction techniques are applied on the images (faces) based on Zernike moments ?and structural similarity measure (SSIM) with local and semi-global blocks. Pre-processing ?is carried out whenever needed, and numbers of measurements are derived. More ?specifically, instead of the usual approach for applying statistics or structural methods ?only, the proposed methodology integrates higher-order representation patterns extracted ?by Zernike moments with a modified version of SSIM (M-SSIM). Individual measurements and metrics resulted from mixed SSIM and Zernike-based approaches give a powerful ?recognition tool with great results. Experiments reveal that correlative Zernike vectors give ?a better discriminant compared with using 2D correlation of the image itself. The ?recognition rate using ORL Database of Faces reaches 98.75%, while using FEI ??(Brazilian) Face Database we got 96.57%. The proposed approach is robust against ?rotation and noise.?

Zernike Moments (ZMs) have been widely used in pattern recognition and image analysis due to their merits of near-zero information redundancy and rotation invariance. However, the discrete sampling error introduced by state-of-the-art algorithms degrades their rotation invariance. In this paper, a novel algorithm for computing the accurate ZMs of binary images is presented. The new algorithm is based on Non-Symmetry Anti-Packing Image Representation (NAIR) and it computes the accurate ZMs of the whole image by summing up the ZMs of all rectangular regions without introducing the discrete sampling error. The experimental results in this paper show that the NAIR-based algorithm for accurate ZMs is faster than the state-of-the-art accurate algorithms.

The aim of our work is to investigate the use of two powerful feature descriptor known as Zernike moments and histogram of oriented gradient (HOG) for facial images extracted from a video sequence uttered ten times by four speakers where MLP neural network classifier is used to identify the correct speaker. We've demonstrated good recognition rate obtained for both Zernike moments and HOG descriptors.