Classification of aerial photograph using neural network

The objective of this paper is to present the development and numerical testing of a robust fault detection and identification (FDI) system using artificial neural networks (ANNs), for incipient (slowly developing) faults occurring in process systems. The challenge in using ANNs in FDI systems arises because of one's desire to detect faults of varying severity, faults from noisy sensors, and multiple simultaneous faults. To address these issues, it becomes essential to have a learning algorithm that ensures quick convergence to a high level of accuracy. A recently developed accelerated learning algorithm, namely a form of an adaptive back propagation (ABP) algorithm, is used for this purpose. The ABP algorithm is used for the development of an FDI system for a process composed of a direct current motor, a centrifugal pump, and the associated piping system. Simulation studies indicate that the FDI system has significantly high sensitivity to incipient fault severity, while exhibiting insensitivity to sensor noise. For multiple simultaneous faults, the FDI system detects the fault with the predominant signature. The major limitation of the developed FDI system is encountered when it is subjected to simultaneous faults with similar signatures. During such faults, the inherent limitation of pattern-recognition-based FDI methodsmore » becomes apparent. Thus, alternate, more sophisticated FDI methods become necessary to address such problems. Even though the effectiveness of pattern-recognition-based FDI methods using ANNs has been demonstrated, further testing using real-world data is necessary.« less

This paper presents a neural network approach to classify three-band RGB color images for automatic visual inspection of seed maize. A back-propagation neural network classifier was developed and tested. The effectiveness of the neural network classification was evaluated by comparing with two conventional statistical methods, minimum distance (MD) and maximum likelihood (ML) classifications. Experimental results showed that the BP neural network classifier, the MD classifier and ML classifier provided the overall accuracies of 93.4%, 91.5% and 96%, respectively. The neural network classifier showed a better performance than the MD and ML classifiers in classifying the shady boundaries or blurred edges of an interested class when the training samples were selected from the boundary areas of the class. This research indicated that neural networks are suitable for the pattern classification of RGB color images.© (1993) COPYRIGHT SPIE--The International Society for Optical Engineering. Downloading of the abstract is permitted for personal use only.

این مطالعه به منظور بررسی پیش بینی شاخص های عملکرد تولیدی در مرغان تخم گذار با استفاده از شبکه های عصبی مصنوعی و رگرسیون غیرخطی چندگانه انجام شد. بررسی بر روی اطلاعات چهار دوره متوالی پرورش در یک واحد پرورش مرغ تخم گذار صورت گرفت. روش های داده-کاوی شامل رگرسیون خطی و غیر خطی، شبکه عصبی پرسپترون سه لایه، شبکه عصبی پرسپترون چهار لایه و شبکه عصبی با تابع پایه ای شعاعی بود. در این مدل ها از متغیرهای سن گله، میزان خوراک مصرفی و فصل تولید به عنوان متغیر پیشگو و شاخص های عملکرد تولیدی شامل درصد تخم-گذاری، وزن توده ای تخم مرغ تولیدی و ضریب تبدیل غذایی به عنوان متغیر پاسخ استفاده شد. نتایج نهایی رگرسیون های خطی نشان داد که برای تمامی متغیرهای وابسته مورد مطالعه متغیر مستقل سن گله معنی دار می باشد. بنابراین رگرسیون غیر خطی شاخص های عملکرد تولیدی در مقابل سن برای مقایسه با شبکه های عصبی مختلف مورد بررسی قرار گرفت و برای مقایسه کلیه مدل ها از ضریب تعیین (R2) و میانگین قدر مطلق خطا (MAE) استفاده شد. نتایج نشان داد بین شبکه های عصبی مصنوعی مختلف مورد مطالعه، شبکه با تابع پایه ای شعاعی بهتر از سایر مدل های در پیش‌بینی شاخص های عملکرد تولیدی مرغان تخم گذار عمل می کند.

A model-based performance test for forest classifiers on remote-sensing imagery

Considering the form diversity of tree species composition in the Bagong Mountain National Forest Park of China, we mapped tree species utilizing Machine Learning Algorithms (support vector machines (SVM) and random forest (RF) classifiers) based on the OHS-2 hyperspectral satellite image by different datasets which combined spectral information and hyperspectral-derived vegetation indices (VIs) for improving tree species classification and explored the best performance of them. To verify the improvement, the results of physically-based spectral classifiers (spectral angle mapper (SAM) and maximum likelihood (ML) classifiers) were applied to compare with the results of machine learning algorithms. The results indicated an overall accuracy of 94.01%, 96.08%, 82.9% and 79.3% for SVM, RF, SAM and ML classifiers of the best performance using different datasets. Highest accuracies resulted from two machine learning algorithms classifiers; SVM and RF compared to SAM and ML classifiers. Although SVM outperformed RF when using all hyperspectral bands and VIs, the overall accuracy of the RF classifier is higher when compared to the SVM classifier using VIs combined selected features. Meanwhile, the RF classifier performed better than SVM after removing the redundancy of spectral data in training samples. Moreover, the machine learning algorithms successfully classified a small number of tree species (Cedrus deodara and Pterocarya stenoptera C. DC.) in the study area, but the physical spectroscopy-based method failed to classify these species. Such integration strategy improved the effectiveness of enhancing the accuracy of tree species classification and mapping their distribution on broad spatial and temporal scales using machine learning algorithms and hyperspectral imagery.

In this paper, a new approach to identify and valid ate the F/A-18 aeroservoelastic model based on flight flutter tests is presented. The Neu ral Network, trained with five different flight flutter cases, is validated using eleven oth er flight flutter test data. The total of sixteen flight flutter tests cases were obtained fo r all three flight regimes (subsonic, transonic and supersonic) at Mach numbers between 0.85 and 1.30 and altitudes between 5,000 feet and 25,000 feet. The obtained r esults highlight the efficiency of the multi-layer perceptron Neural Network in model identification. The Neural Network optimization is required mixing hidden layer size r eduction and four-layered Neural Network performances. This article shows that a four-layered Neural Network with only 16 neurons is sufficient to create an accurate mode l. The fit coefficients are higher than 92%, either for the identification test data or the validation ones, and thus the Neural Network accuracy was demonstrated.

An algorithm of dynamic watermark based on neural network is presented which is more robust against attack of false authentication and watermark-tampered operations contrasting with one watermark embedded method. (1) Five binary images used as watermarks are coded into a binary array. The total number of 0s and 1s is 5*N, every 0 or 1 is enlarged fivefold by information-enlarged technique. N is the original total number of the watermarks' binary bits. (2) Choose the seed image pixel p_x,y and its 3&times;3 vicinities pixel p _x-1,y-1,p_x-1,y,p_x-1,y+1,p_x,y-1,p_x,y+1,p_x+1,y-1,p_x+1,y,p_x+1,y+1 as one sample space. The p_x,y is used as the neural network target and the other eight pixel values are used as neural network inputs. (3) To make the neural network learn the sample space, 5*N pixel values and their closely relevant pixel values are randomly chosen with a password from a color BMP format image and used to train the neural network.(4) A four-layer neural network is constructed to describe the nonlinear mapped relationship between inputs and outputs. (5) One bit from the array is embedded by adjusting the polarity between a chosen pixel value and the output value of the model. (6) One randomizer generates a number to ascertain the counts of watermarks for retrieving. The randomly ascertained watermarks can be retrieved by using the restored neural network outputs value, the corresponding image pixels value, and the restore function without knowing the original image and watermarks (The restored <i>coded-watermark</i>_bit=1, if <i>ox,y(restored)</i>&gt;p_x,y(<i>reconstructed, else coded-watermark_bit </i>=0). The retrieved watermarks are different when extracting each time. The proposed technique can offer more watermarking proofs than one watermark embedded algorithm. Experimental results show that the proposed technique is very robust against some image processing operations and JPEG lossy compression. Therefore, the algorithm can be used to protect the copyright of one important image.

This is an extension of previous work which used an artificial neural network with a back-propagation algorithm and a lookup table to find the inverse kinematics for a manipulator arm moving along pre-defined trajectories. The work now described shows that the performance of this technique can be improved if the back-propagation is made to be adaptive. Also, further improvement is obtained by using the whole workspace to train the neural network rather than just a pre-defined path. For the inverse kinematics of the whole workspace, a comparison has also been done between the adaptive back-propagation algorithm and radial basis function.

Maximum likelihood classifier that is often used in classification of satellite images assumes the distribution of each class to Gaussian. Such linear classifier can classify correctly when the case that classification probability of each class is exclusive. Remotely sensed data, however, belong to several classes and have non-linear separable condition. To improve the classification accuracy of non-linear separable data, the application of the single-step multi-layer back propagation neural networks have been studied by many researchers. In this paper, multi-step multi-layer neural networks, so called cooperative learning neural networks, are proposed to classify the non-linear separable satellite data.The cooperative learning neural network consists of extraction networks for each class and an unification network which unifies the extracted values. The unification network is also used for unification of different environments such as time-series data or neighboring regions. The result of the classification of LANDSAT TM data of Nagoya city using the cooperative learning neural network is introduced. Classified image is compared with the detailed digital land cover information (TDT-112) and the images classified using single-step multi-layer neural network, maximum likelihood classifier and fuzzy set reasoning. As the result of the comparison, the cooperative learning neural network classify the remote sensing data more exactly than the other methods.

Optical and SAR sensor synergies for forest and land cover mapping in a tropical site in West Africa

ABSTRACTAlthough satellite remote sensing imagery is suitable for mapping and monitoring the surface water, the application of surface water classification techniques in the literatures is still constrained by low accuracy in various situations. As the GF-4 satellite is one of the few moderate spatial resolution geostationary orbit remote sensing satellites, the purpose of this study is to introduce a method that consistently improves the accuracy of surface water classification by utilizing the temporal characteristics of the dense time series of GF-4 images. A new time series water index called the spectrum and solar altitude angle water index (SSWI) is proposed based on the characteristics of the variation in the top of atmosphere (TOA) radiance with the change in solar altitude angle in the near-infrared (NIR) band of GF-4 images. The thresholding technique is applied to the SSWI map to automatically extract surface water. The performance of the SSWI method for surface water classification is validated and compared with the results of two widely used water classification methods, the Normalized Difference Water Index (NDWI) and maximum likelihood (ML) classifier, at four representative sites in different parts of China, including Beijing, Hubei, Tibet and Guangdong. The results show that at all four test sites, the classification accuracy of SSWI is significantly higher than that of NDWI and ML. Averaged over the four test sites, the total error of the SSWI for water bodies is only approximately 36.9% of that of the NDWI method and 23.5% of that of the ML classifier. In particular, the SSWI method shows outstanding performance in terms of distinguishing surface water from backgrounds that are usually sources of surface water classification errors, e.g. mountain shadows and low-albedo built-up areas. Therefore, the SSWI method can be used to classify surface water from GF-4 images and holds the potential to be a useful surface water classification technology for water resource studies and applications.

Deals with the approximation of continuous functions by feedforward neural networks. After presenting one of the main results of Ito, the paper tries to get a universal approximator implementable as a (1+1/2) layer neural network using Heaviside functions as univariate functions. The paper presents an explicit formula for function approximation implementable as a three-layer feedforward neural network instead of a four-layer neural network. These three-layer feedforward neural networks have the same number of neurons in the hidden layer as the equivalent four-layer neural networks have in the second hidden layer.

Recently, classifying the modulation schemes of signals using deep neural network has received much attention. In this paper, we introduce a general model of deep neural network (DNN)-based modulation classifiers for single-input single-output (SISO) systems. Its feasibility is analyzed using maximum a posteriori probability (MAP) criterion and its robustness to uncertain noise conditions is compared to that of the conventional maximum likelihood (ML)-based classifiers. To reduce the design and training cost of DNN classifiers, a simple but effective pre-processing method is introduced and adopted. Furthermore, featuring multiple recurrent neural network (RNN) layers, the DNN modulation classifier is realized. Simulation results show that the proposed RNN-based classifier is robust to the uncertain noise conditions, and the performance of it approaches to that of the ideal ML classifier with perfect channel and noise information. Moreover, with a much lower complexity, it outperforms the existing ML-based classifiers, specifically, expectation maximization (EM) and expectation conditional maximization (ECM) classifiers which iteratively estimate channel and noise parameters. In addition, the proposed classifier is shown to be invariant to the signal distortion such as frequency offset. Furthermore, the adopted pre-processing method is shown to accelerate the training process of our proposed classifier, thus reducing the training cost. Lastly, the computational complexity of our proposed classifier is analyzed and compared to other traditional ones, which further demonstrates its overall advantage.

The design, implementation, and preliminary testing of a computer system for automatic multispectral magnetic resonance imaging analysis is presented. The modular structure of the system permits easy comparison between various classification algorithms. The classification accuracy of traditional statistical pattern-recognition algorithms is compared to the results that can be obtained with neural networks of different topologies. Quantitative (confusion matrices) as well as visual (segmented images) results of a study performed on sets of normal and pathological images are presented. Images segmented with a neural network classifier (NNC) appear less noisy than images segmented with a maximum likelihood classifier (MLC), and it has been observed that the NNC is less sensitive to the selection of the training sets than the MLC

Ground cover classification based on a single satel- lite image can be challenging. The work reported here concerns the use of multitemporal satellite image data in order to alleviate this problem. We consider the problem of vegetation mapping and model the phenological evolution of the vegetation using a Hidden Markov Model (HMM). The different vegetation classes can be in one of a predefined set of states related to their phenological development. The characteristics of a given class are specified by the state transition probabilities as well as the probability of given satellite observations for that class and state. Classification of a specific pixel is thus reduced to selecting the class that has the highest probability of producing a given series of observations for that pixel. Compared to standard classification techniques such as maximum likelihood (ML) classification, the proposed scheme is flexible in that it derives its properties not only from image specific training data, but also from a model of the temporal behavior of the ground cover. It is shown to produce results that compare favorably to those obtained using ML classification on single satellite images, it also generalizes better than this approach. Obtaining good ground cover classifications based on a single satellite image can be challenging. The work reported here concerns the use of multitemporal satellite image data in order to alleviate this problem. We will consider an application of these methods to mapping of high mountain vegetation in Norway. The traditional mapping method based on manual field work is prohibitively expensive and alternatives are therefore sought. Vegetation classification based on satellite images is an interesting alternative, but the complexity of the vegetation ground cover is high and the use of multitemporal satellite image acquisitions is shown to improve the classifi- cation quality. This document is organized as follows: In the next section, we briefly recapitulate previous work related to multitemporal satellite image classification and phenological models. In section IV we discuss the HMM and how it can be used for classification. In section V we show results of the application of our algorithm, conclusions are given in section VI.