Array Normalization Algorithms Applied to Qualitative Electronic Noses

2013 ◽  
Vol 303-306 ◽  
pp. 284-287
Author(s):  
Zhou Ye Chen ◽  
Lei He
Keyword(s):  
Electronic Nose ◽  
Recognition Rate ◽  
Correct Recognition ◽  
Training Dataset ◽  
Electronic Noses ◽  
Array Normalization ◽  
Negative Effect ◽  
Correct Recognition Rate ◽  
The Impact ◽  
Better Than

In a qualitative electronic nose, different gas concentrations of the training dataset will have a negative effect on the correct recognition rate of the system. In order to reduce or eliminate the impact of the factor of concentration on the qualitative electronic nose, array normalization algorithms are proposed. In this paper, six different array normalization algorithms were studied and compared in different application cases. All of these algorithms are effective in increasing the correct recognition rate of the qualitative electronic nose and different algorithms are biased in favor of different application directions. The algorithms I II and III are most commonly used ones because of their stableness, the algorithms with global compression are better than the ones with local compression when more sensors are used in a array.

scholarly journals IDENTIFIKASI IRIS MENGGUNAKAN IMPROVED LEVENBERG-MARQUARDT

2015 ◽  
Vol 1 (1) ◽  
pp. 10
Author(s):  
Rocky Yefrenes Dillak
Keyword(s):  
Hough Transform ◽  
Recognition Rate ◽  
Correct Recognition ◽  
Gray Level ◽  
Maximum Probability ◽  
Circular Hough Transform ◽  
Marquardt Algorithm ◽  
Levenberg Marquardt ◽  
Correct Recognition Rate

Sistem biometrika adalah suatu sistem pengenalan diri menggunakan bagian tubuh atau perilaku manusia seperti sidik jari, telapak tangan, telinga, retina, iris mata, wajah, suhu tubuh, tanda tangan, dll. Iris mata merupakan salah satu biometrika yang sangat stabil, handal, akurat dan merupakan metode autentikasi biometrika tercepat  oleh karena itu merupakan suatu topik penelitian yang sangat diminati oleh banyak peneliti. Penelitian ini bertujuan untuk mengembangkan suatu metode yang dapat digunakan untuk mengidentifikasi secara otomatis seseorang berdasarkan citra iris mata miliknya menggunakan jaringan syaraf tiruan levenberg-marquardt. Penelitian ini menggunakan metode deteksi tepi cany dan circular hough transform untuk segmentasi wilayah iris yang terletak diantara pupil dan sclera serta metode ekstraksi ciri gray level cooccurence matrix (GLCM) yang digunakan untuk ekstraksi ciri. Ciri-ciri tersebut adalah maximum probability, correlation, contrast, energy, homogeneity, dan entropy. Ciri-ciri tersebut kemudian dilatih menggunakan jaringan syaraf tiruan dengan aturan pembelajaran levenberg–marquardt algorithm untuk mengidentifikasi seseorang berdasarkan citra irisnya. Penelitian ini menggunakan 150 data citra iris yang masing-masing terbagi atas 100 data citra iris untuk pelatihan dan 50 data citra iris  untuk pengujian. Berdasarkan hasil pengujian yang dilakukan diperoleh correct recognition rate (CRR) sebesar 99.98%  yang menunjukkan bahwa metode ini dapat digunakan untuk mengidentifikasi secara otomatis seseorang berdasarkan citra iris mata miliknya.

scholarly journals Active Learning for Recognition of Shipwreck Target in Side-Scan Sonar Image

2019 ◽  
Vol 11 (3) ◽  
pp. 243 ◽  
Author(s):  
Bangyan Zhu ◽  
Xiao Wang ◽  
Zhengwei Chu ◽  
Yi Yang ◽  
Juan Shi
Keyword(s):  
Feature Extraction ◽  
Target Recognition ◽  
Recognition Rate ◽  
Correct Recognition ◽  
Matching Model ◽  
Side Scan Sonar ◽  
Recognition Model ◽  
Adaboost Algorithm ◽  
Recognition Process ◽  
Correct Recognition Rate

In order to realize the automatic and accurate recognition of shipwreck targets in side-scan sonar (SSS) waterfall images, a pipeline that contains feature extraction, selection, and shipwreck recognition, an AdaBoost model was constructed by sample images. Shipwreck targets are detected quickly by a nonlinear matching model, and a shipwreck recognition in SSS waterfall images are given, and according to a wide set of combinations of different types of these individual procedures, the model is able to recognize the shipwrecks accurately. Firstly, two feature-extraction methods suitable for recognizing SSS shipwreck targets from natural sea bottom images were studied. In addition to these two typical features, some commonly used features were extracted and combined as comprehensive features to characterize shipwrecks from various feature spaces. Based on Independent Component Analysis (ICA), the preferred features were selected from the comprehensive features, which avoid dimension disaster and improved the correct recognition rate. Then, the Gentle AdaBoost algorithm was studied and used for constructing the shipwreck target recognition model using sample images. Finally, a shipwreck target recognition process for the SSS waterfall image was given, and the process contains shipwreck target fast detection by a nonlinear matching model and accurate recognition by the Gentle AdaBoost recognition model. The results show that the correct recognition rate of the model for the sample image is 97.44%, while the false positive rate is 3.13% and the missing detection rate is 0. This study of a measured SSS waterfall image confirms the correctness of the recognition process and model.

Feature Extraction by Decryption to Target

2014 ◽  
Vol 1008-1009 ◽  
pp. 1509-1512
Author(s):  
Qing E Wu ◽  
Hong Wang ◽  
Li Fen Ding
Keyword(s):  
Feature Extraction ◽  
Wavelet Transform ◽  
Extraction Method ◽  
Recognition Rate ◽  
Region Of Interest ◽  
Correct Recognition ◽  
Palmprint Recognition ◽  
Feature Extraction Method ◽  
Energy Feature ◽  
Correct Recognition Rate

To carry out an effective classification and recognition for target, this paper studied the target owned characteristics, discussed a decryption algorithm, gave a feature extraction method based on the decryption process, and extracted the feature of palmprint in region of interest. Moreover, this paper used the wavelet transform to extract the energy feature of target, gave an approach on matching and recognition to improve the correctness and efficiency of existing recognition approaches, and compared it with existing approaches of palmprint recognition by experiments. The experiment results show that the correct recognition rate of the approach in this paper is improved averagely by 2.34% than that of the existing recognition approaches.

Is something better than nothing? The impact of foreclosed and lease-purchase properties on residential property values

Urban Studies ◽  
2017 ◽  
Vol 54 (16) ◽  
pp. 3681-3699 ◽  
Author(s):  
Youngme Seo ◽  
Michael Craw
Keyword(s):  
Affordable Housing ◽  
Property Values ◽  
High Income ◽  
Residential Property ◽  
Negative Effect ◽  
Residential Property Values ◽  
The Impact ◽  
Better Than ◽  
Cleveland Ohio

Lease-purchase (L-P) programmes that rehabilitate foreclosed property for sale as affordable housing may provide a way to reduce foreclosure externalities on nearby property values. This paper investigates the feasibility of such a strategy by estimating the effects of foreclosed properties on nearby residential property values compared with those of an L-P programme operated by the Cleveland Housing Network, Cleveland, Ohio. The findings indicate that although both L-P and foreclosed properties have a negative effect on the value of nearby non-distressed homes, the negative effect of foreclosure is larger. At the same time, the scope of the foreclosure externality is greater in low- and moderate-income neighbourhoods, while the foreclosure externality is generally smaller in high income neighbourhoods. Such results imply that an L-P strategy is likely to be more effective in offsetting foreclosure externalities in low- and moderate-income neighbourhoods than in high income neighbourhoods.

Study on face recognition under unconstrained conditions based on LBP and deep learning

2020 ◽  
pp. 1-12
Author(s):  
Liping Zhou ◽  
Mingwei Gao ◽  
Chun He
Keyword(s):  
Deep Learning ◽  
Face Recognition ◽  
Input Data ◽  
Recognition Rate ◽  
Texture Feature ◽  
Recognition Algorithm ◽  
Correct Recognition ◽  
Layer By Layer ◽  
The Face ◽  
Correct Recognition Rate

At present, the correct recognition rate of face recognition algorithm is limited under unconstrained conditions. To solve this problem, a face recognition algorithm based on deep learning under unconstrained conditions is proposed in this paper. The algorithm takes LBP texture feature as the input data of deep network, and trains the network layer by layer greedily to obtain optimized parameters of network, and then uses the trained network to predict the test samples. Experimental results on the face database LFW show that the proposed algorithm has higher correct recognition rate than some traditional algorithms under unconstrained conditions. In order to further verify its effectiveness and universality, this algorithm was also tested in YALE and YALE-B, and achieved a high correct recognition rate as well, which indicated that the deep learning method using LBP texture feature as input data is effective and robust to face recognition.

scholarly journals Improved Faster RCNN Based on Feature Amplification and Oversampling Data Augmentation for Oriented Vehicle Detection in Aerial Images

2020 ◽  
Vol 12 (16) ◽  
pp. 2558 ◽  
Author(s):  
Nan Mo ◽  
Li Yan
Keyword(s):  
Data Augmentation ◽  
Vehicle Detection ◽  
Aerial Images ◽  
Training Dataset ◽  
Discriminative Ability ◽  
Feature Map ◽  
Negative Effect ◽  
Bounding Boxes ◽  
The Impact ◽  
Oriented Bounding Boxes

Vehicles in aerial images are generally with small sizes and unbalanced number of samples, which leads to the poor performances of the existing vehicle detection algorithms. Therefore, an oriented vehicle detection framework based on improved Faster RCNN is proposed for aerial images. First of all, we propose an oversampling and stitching data augmentation method to decrease the negative effect of category imbalance in the training dataset and construct a new dataset with balanced number of samples. Then considering that the pooling operation may loss the discriminative ability of features for small objects, we propose to amplify the feature map so that detailed information hidden in the last feature map can be enriched. Finally, we design a joint training loss function including center loss for both horizontal and oriented bounding boxes, and reduce the impact of small inter-class diversity on vehicle detection. The proposed framework is evaluated on the VEDAI dataset that consists of 9 vehicle categories. The experimental results show that the proposed framework outperforms previous approaches with a mean average precision of 60.4% and 60.1% in detecting horizontal and oriented bounding boxes respectively, which is about 8% better than Faster RCNN.

scholarly journals Regulatory Environment and Firm Performance in EU Telecommunications Services

2014 ◽  
Vol 13 (3) ◽  
Author(s):  
Daniel Montolio ◽  
Francesc Trillas ◽  
Elisa Trujillo-Baute
Keyword(s):  
Firm Performance ◽  
Market Share ◽  
Multinational Firms ◽  
Country Level ◽  
Level Information ◽  
Negative Effect ◽  
Telecommunications Services ◽  
Broadband Market ◽  
The Impact ◽  
Better Than

AbstractWe empirically estimate the effects of regulated access prices and firms’ multinational status on firm performance by using firm, corporate group, and country level information for the European broadband market between 2002 and 2010. Three measures of firm performance are used, namely: market share, revenue and productivity. Special attention is paid to differences in the impact on the performance measures depending on a firm’s position as either a market incumbent or entrant. We find that while access prices have a negative effect on entrants’ market share and revenue, the effect on incumbents’ market share, revenue and productivity is positive. Further, we find that multinational entrants perform better than national entrants in terms of their market share but worse in terms of their revenue and productivity. The opposite is true of incumbent multinationals which perform better than nationals in terms of their revenue and productivity but worse in terms of their market share. This confirms that a firm’s multinational status has a significant impact on its performance, and that this impact differs for incumbents and entrants. Finally, when evaluating the impact of access prices on firm performance at the mean performance of national and multinational firms, we find that the impact of access prices is lower for multinational than for national firms.

scholarly journals The Use of Fuzzy BackPropagation Neural Networks for the Early Diagnosis of Hypoxic Ischemic Encephalopathy in Newborns

2011 ◽  
Vol 2011 ◽  
pp. 1-5 ◽  
Author(s):  
Liu Li ◽  
Huo Liqing ◽  
Lu Hongru ◽  
Zhang Feng ◽  
Zheng Chongxun ◽  
...  
Keyword(s):  
Neural Networks ◽  
Artificial Neural Networks ◽  
Early Diagnosis ◽  
Recognition Rate ◽  
Hypoxic Ischemic Encephalopathy ◽  
Correct Recognition ◽  
Backpropagation Neural Networks ◽  
Artificial Neural ◽  
Correct Recognition Rate ◽  
Ischemic Encephalopathy

Objective. To establish an early diagnostic system for hypoxic ischemic encephalopathy (HIE) in newborns based on artificial neural networks and to determine its feasibility.Methods. Based on published research as well as preliminary studies in our laboratory, multiple noninvasive indicators with high sensitivity and specificity were selected for the early diagnosis of HIE and employed in the present study, which incorporates fuzzy logic with artificial neural networks.Results. The analysis of the diagnostic results from the fuzzy neural network experiments with 140 cases of HIE showed a correct recognition rate of 100% in all training samples and a correct recognition rate of 95% in all the test samples, indicating a misdiagnosis rate of 5%.Conclusion. A preliminary model using fuzzy backpropagation neural networks based on a composite index of clinical indicators was established and its accuracy for the early diagnosis of HIE was validated. Therefore, this method provides a convenient tool for the early clinical diagnosis of HIE.

scholarly journals Development of a LeNet-5 Gas Identification CNN Structure for Electronic Noses

Sensors ◽  
2019 ◽  
Vol 19 (1) ◽  
pp. 217 ◽  
Author(s):  
Guangfen Wei ◽  
Gang Li ◽  
Jie Zhao ◽  
Aixiang He
Keyword(s):  
Neural Network ◽  
Neural Networks ◽  
Electronic Nose ◽  
Probabilistic Neural Network ◽  
Recognition Rate ◽  
Test System ◽  
Identification Accuracy ◽  
System Response ◽  
Electronic Noses ◽  
Gas Identification

A new LeNet-5 gas identification convolutional neural network structure for electronic noses is proposed and developed in this paper. Inspired by the tremendous achievements made by convolutional neural networks in the field of computer vision, the LeNet-5 was adopted and improved for a 12-sensor array based electronic nose system. Response data of the electronic nose to different concentrations of CO, CH4 and their mixtures were acquired by an automated gas distribution and test system. By adjusting the parameters of the CNN structure, the gas LeNet-5 was improved to recognize the three categories of CO, CH4 and their mixtures omitting the concentration influences. The final gas identification accuracy rate reached 98.67% with the unused data as test set by the improved gas LeNet-5. Comparison with results of Multiple Layer Perceptron neural networks and Probabilistic Neural Network verifies the improvement of recognition rate while with the same level of time cost, which proved the effectiveness of the proposed approach.

A Recognition Algorithm for M-QAM Based on Amplitude Distribution

2011 ◽  
Vol 128-129 ◽  
pp. 20-24
Author(s):  
Lu Yuan Tan ◽  
Qian Wang ◽  
Xiao Yan ◽  
Kai Yu Qin
Keyword(s):  
Carrier Frequency ◽  
Carrier Frequency Offset ◽  
Recognition Rate ◽  
Amplitude Distribution ◽  
Frequency Offset ◽  
Recognition Algorithm ◽  
Automatic Recognition ◽  
Correct Recognition ◽  
Nonlinear Amplification ◽  
Correct Recognition Rate

An automatic recognition algorithm for M-QAM based on the amplitude distribution is proposed. This algorithm uses the normalized amplitude distribution to achieve automatic recognition for M-QAM signals, and enhances the correct recognition rate through the nonlinear amplification. Compared with the recognition algorithm based on amplitude moment, this algorithm does not need those prior conditions, such as carrier frequency offset, code element rate, amplitude factor and so on. The simulation confirmed that, when SNR≥16dB the correct recognition rate of this algorithm is greater than 90%.

Sign in / Sign up

Export Citation Format

Share Document