Large-scale, high-resolution imaging of cerebral hemodynamics is essential for brain research. Uniquely capable of comprehensive quantification of cerebral hemodynamics and oxygen metabolism in rodents based on the endogenous hemoglobin contrast, multiparametric photoacoustic microscopy (PAM) is ideally suited for this purpose. However, the out-of-focus issue due to the uneven surface of the rodent brain results in inaccurate PAM measurements and presents a significant challenge to cortex-wide multiparametric recording. We report a large-scale, high-resolution, multiparametric PAM system based on real-time surface contour extraction and scanning, which avoids the prescan and offline calculation of the contour map required by previously reported contour-scanning strategies. The performance of this system has been demonstrated in both phantoms and the live mouse brain through a thinned-skull window. Side-by-side comparison shows that the real-time contour scanning not only improves the quality of structural images by addressing the out-of-focus issue but also ensures accurate measurements of the concentration of hemoglobin (CHb), oxygen saturation of hemoglobin (sO2), and cerebral blood flow (CBF) over the entire mouse cortex. Furthermore, quantitative analysis reveals how the out-of-focus issue impairs the measurements of CHb, sO2, and CBF.
Encoder-decoder framework attracts great interests in image caption. It focuses on the extraction of low-level features and achieves good results. The performance can be further improved if high-level semantics are considered. In this work, we propose a new image caption model incorporating high-level semantic features through an revised Convolutional Neural Network(CNN). Both the low-level image features and high-level semantic features are fed into the Long-Short Term Memory networks(LSTMs) to acquire natural sentence descriptions. We show in a number of experiments on Flickr8K and Flickr30K datasets that our method outperforms most standard network baseline for image caption.
Convolutional neural networks in deep learning models have dominated the recent image recognition works. But the lack of capacity to maintain spatial invariance makes identification of micronucleus cells as a classic task in digital pathology still a challenge task. In this paper, a novel convolutional neural network for feature maps spatial transformation (FSTCNN) is proposed, which incorporates a Spatial Transformer Network. Our model allows the spatial manipulation of data within the network, provides the ability of active spatial transformation for neural network without any extra supervision. We compared the results of inserting STN into different convolutional layers and found that such a network can transform the input image more steadily, correct the image to one certain position, make it fill the whole screen to create a better environment for image recognition. The results show a distinct advantage over other convolutional neural networks for medical image recognition.
The development of convolutional neural network has brought great achievements to image classification in recent years. However, the classification performance is good only for natural images rather than medical images. An important reason is that the medical image database used for training is always deficient. So how to use these limited data to acquire more extensive features has become a hot research focus. In this paper, we first update the order and number of the whole training data every time in active and incremental fine-tuning. Then we set different contribution rate for the data selected in our model, which based on the information quantity of the data in training stage and make our model converge steadily. After that, a pre-trained model and our preprocessed datasets are employed, which allows us to further fine-tune our models. The experiments evaluated on two different biomedical datasets shows that our model can achieve promising results.
Due to the advantages of having large storage capacity and small code area, QR (quick response) codes have been widely used for automatic identification in many commercial applications such as parcel packaging, business cards and etc. The existing methods mainly focus on unambiguous QR code location with simple background, which always rely on the accomplishment of machine independently. While the QR code images with low quality and complex background always affect the accuracy and efficiency of location in automatic identification, especially the QR code images in which the finder patterns are destroyed. With the help of human, many interactive learning approaches can solve the problem of cognitive obstacles in computer operations. This paper focuses on locating blur QR codes with complex background by an efficient interactive two-stage framework. The first stage is rough location, which includes our interactive feature template setting and clustering process with our improved mean shift algorithm. Then we do the accurate location based on the optimization of the finder pattern detection. Experiments are performed on damaged, contaminated and scratched images with a complex background, which provide a quite promising result for QR code location.
Automatic image annotation is now a tough task in computer vision, the main sense of this tech is to deal with managing the massive image on the Internet and assisting intelligent retrieval. This paper designs a new image annotation model based on visual bag of words, using the low level features like color and texture information as well as mid-level feature as SIFT, and mixture the pic2pic, label2pic and label2label correlation to measure the correlation degree of labels and images. We aim to prune the specific features for each single label and formalize the annotation task as a learning process base on Positive-Negative Instances Learning. Experiments are performed using the Corel5K Dataset, and provide a quite promising result when comparing with other existing methods.
With the development of barcodes for commercial use, people’s requirements for detecting barcodes by smart phone become increasingly pressing. The low quality of barcode image captured by mobile phone always affects the decoding and recognition rates. This paper focuses on locating and decoding EAN-13 barcodes in fuzzy images. We present a more accurate locating algorithm based on segment length and high fault-tolerant rate algorithm for decoding barcodes. Unlike existing approaches, location algorithm is based on the edge segment length of EAN -13 barcodes, while our decoding algorithm allows the appearance of fuzzy region in barcode image. Experimental results are performed on damaged, contaminated and scratched digital images, and provide a quite promising result for EAN -13 barcode location and decoding.
The order of cigarette market is a key issue in the tobacco business system. The anti-counterfeiting code, as a kind of effective anti-counterfeiting technology, can identify counterfeit goods, and effectively maintain the normal order of market and consumers' rights and interests. There are complex backgrounds, light interference and other problems in the anti-counterfeiting code images obtained by the tobacco recognizer. To solve these problems, the paper proposes a locating method based on Susan operator, combined with sliding window and line scanning,. In order to reduce the interference of background and noise, we extract the red component of the image and convert the color image into gray image. For the confusing characters, recognition results correction based on the template matching method has been adopted to improve the recognition rate. In this method, the anti-counterfeiting code can be located and recognized correctly in the image with complex background. The experiment results show the effectiveness and feasibility of the approach.
An identical weak reflection FBGs demodulation system based on a FDML laser is proposed. The laser is developed to output a continuous wavelength-swept spectrum in the scanning frequency of 120 kHz over a spectral range of more than 10nm at 1.54 μm. Based on this high-speed wavelength-swept light and the optical transmission delay effect, the demodulation system obtains the location and wavelength information of all identical weak FBGs by the reflected spectrum within each scanning cycle. By accessing to a high-speed FPGA processing module, continuous demodulation of 120 kHz is realized. The system breakthroughs the bandwidth of the laser to expand the sensors capacity and greatly improves the demodulation speed of a TDM sensing network. The experiments show the system can distinguish and demodulate the identical weak FBGs and measure the 4 kHz vibration at 120 kHz demodulation speed.
In this paper, a new means of measuring the displacement of GMA (Giant Magnetostrictive Actuator) based on FGB (Fiber Bragg Grating) sensor is proposed, experimental results confirmed that FGB sensor can measure the displacement of GMA in different frequencies and achieve good results. In addition a modified Bouc-Wen model is presented to describe the GMA, the proposed model can describe the asymmetric hysteresis of GMA from 1 Hz to 100 Hz well, and DE(Differential Evolution) algorithm is used for adaptive identification of the GMA system, the algorithm has fast convergence and high accuracy. Finally, it verifies that the identification model fits the experimental data well.
With the development of industry and agriculture, the environmental pollution becomes more and more serious. Various kinds of poisonous gas are the important pollution sources. Various kinds of poisonous gas, such as the carbon monoxide, sulfureted hydrogen, sulfur dioxide, methane, acetylene are threatening human normal life and production seriously especially today when industry and various kinds of manufacturing industries develop at full speed. The acetylene is a kind of gas with very lively chemical property, extremely apt to burn, resolve and explode, and it is great to destroy things among these poisonous gases. Comparing with other inflammable and explosive gas, the explosion range of the acetylene is heavier. Therefore carrying on monitoring acetylene pollution sources scene in real time, grasping the state of pollution taking place and development in time, have very important meanings. Aim at the above problems, a set of optical fiber detection system of acetylene gas based on the characteristic of spectrum absorption of acetylene is presented in this paper, which has reference channel and is for on-line and real-time detection. In order to eliminate the effect of other factors on measurement precision, the double light sources, double light paths and double cells are used in this system. Because of the use of double wavelength compensating method, this system can eliminate the disturbance in the optical paths, the problem of instability is solved and the measurement precision is greatly enhanced. Some experimental results are presented at the end of this paper.
Sensor technology is one of three major pillars of the modern information technology. With the extensive application of sensor, the dependability of the sensor is paid more and more attention. The development of sensor faults diagnose technology offers strong guarantee for using the sensor reliably. In this paper, the application of combining the wavelet and BP neural networks to sensors failure detection is studied, and a novel diagnosis method based on discrete wavelet transform and BP neural network was proposed to detect and identify sensor abrupt fault. Since wavelet transform can accurately localize sensor signal characteristics both in time and frequency domain, it is very suitable for non-stationary signal analysis. After discrete wavelet transform analysis for sensor output, eigenvector of energy changing rate was extracted, and classification of sensor fault was conducted by using BP neural network. The proposed method does not need construction of sensor model and measurement of sensor input. Hence redundant data can be reduced by omitting some wavelet coefficients and the capability of fault detection can be improved. Sensor fault diagnosis is simulated by the computer. Through a large amount of simulated examples it indicates that the sensors fault diagnosis method based on the theory of wavelet has characteristic such as good sensitivity, high accuracy rate and robust ability to overcome noise. Simulation results proved the effectiveness of this method.
The purpose of this paper is to present a novel three-layer adaptive multisensor data fusion system, which is appropriate to the harsh environment. In order to overcome the noise in the data collected by the sensors in the harsh environment, the first layer of the system is the data pretreatment layer. In this layer, the data collected by the sensor array is denoised by the wavelet threshold algorithm, which provides reliable data to the next data fusion Layer. Taking use of the good error tolerance and self-studying performance of NN (neural network), the data from the first layer is fused by the second layer--- data fusion layer based on NN. The third layer is the feedback layer, in which the output signal is feedback to the second layer. The adaptive algorithm will adjust the weights of the units in the NN, which implements the adaptive ability of the whole system. The experimental results presented in the paper indicate that the system proposed here implements data fusion effectively, its fusion precision is improved compared with the traditional fusion system, and has many advantages like strong adaptive ability, high SNR (signal-to-noise ratio) and low distortion, etc.
The elliptic curve cryptographic random sequences as watermark are embedded in wavelet transform domain of the cover image. This algorithm takes advantages of the multiresolution feature of wavelet transform and non-relevant feature of the cryptographic random signal. The cryptographic random sequences are generated by the elliptic curve group and Galois Field function selected. The experimental results demonstrate that the scheme proposed is security, invisible and robust against commonly image processing techniques.
With the rapid globalization of market and business, E-trading affects every manufacture enterprise. However, the security of network manufacturing products of transmission on Internet is very important. In this paper we discussed the protocol of fair exchange and platform for network manufacture products E-trading based on fair exchange protocol and digital watermarking techniques. The platform realized reliable and copyright protection.
When enjoying videophone or distant learning, people want to see human face as real as possible even in very low bit rate. How to synthesis human face to deliver over network such as Internet and PSTN draws much attention. Conventional techniques based on low-level features cannot perform the desired operation. While model based method need much prior knowledge. The authors present a new algorithm for human face synthesis. It can give a virtual face based on human vision system for bit rate ranging from several kb/s to tens of KB/s. An Adaptive Face Image Filter(AFIF) is used to attenuate noise and preserve face edges as well as details. A facial region detection method detects those pixels that belong to a face. After that, with a novel facial texture interpolating method, the face is rendered in gray scale. Its key feature is a group of diffuse functions for interpolation. Then color is rendered to the whole face scalable.
In this paper the authors study the effect of temperature in magneto-optical disk recording and the importance of the disk temperature measuring. We design the sensor system for the control and measure of magneto-optical disk temperature. The system is well received by users because of its perfect function in usage software and user interface.
The application of sensors cover over all the fields, including signal detection. To meet the increased requirements, many new sensors and their corresponding technologies have been exploited. Micro sensor, smart sensor, and multifunction sensor are reported in this paper. The three kinds of sensors are the typical new sensors and represent the main trend of the sensors of the next century. The paper also describes a number of succeeded applications in signal detection as examples.