Unmanned Aerial Vehicles(UAVs) are widely used for civilian and military purposes, such as surveillance, reconnaissance, search and rescue, border patrol etc. In this paper a single UAV is equipped with a gimballed camera to conduct surveillance operations, the route planning method that uses neural...

In this paper, an optimization method of wavelet neural network for structure damage identification is established. The suspension bridge is used as the research object. Firstly, wavelet coefficients modulus maxima are obtained by wavelet transform to determine the location of structural damage. Then,...

This paper proposes a fast multi-domain convolutional neural networks method (Fast MDNet) for visual tracking. Fast MDNet builds on fast region-based convolutional neural networks (Fast R-CNN) and MDNet to efficiently track arbitrary objects using deep convolutional networks. We introduce a RoI pooling...

Convolution operation is the most important and time consuming step in a convolution neural network model. In this work, we analyze the computing complexity of direct convolution and fast-Fourier-transform-based (FFT-based) convolution. We creatively propose CS-unit, which is equivalent to a combination...

Localization estimation of sensor node is a key component in many sensor networks' applications. Correct localization is required for WSN applications, but it is often expensive to include GPS adapters in each sensor node. The RSSI ranged-based localization algorithm is a simple and cost effective localization...

Gender information, as a crucial part of human demographics, is valuable for its abundant connotations and potential applications. Though much effort has been made on the problem of gender inference, most existing methods are highly dependent on data from specific sources, like Twitter, and are difficult...

Traffic flow prediction have the characteristics of complexity, non-linearity and randomness. Researchers are resorting to hybrid neural networks fusing more effective algorithms into forecast process. This paper advances a fresh algorithm with both the thought of variation and adaptive to conquer the...

It is an open question if the explosive computing capability of artificial intelligence could transcend human intelligence. It is proposed that the universe is bisected and encapsulated into Self and Outside World by an individual based on one's sensitive tactility acquired in the biological evolution...

Low-resolution face recognition (LR FR) has become an active research subarea due to its significances for real applications. Conventional low-resolution face recognition approaches meet challenges like noise affection and lack of effective features with LR faces. In this paper, we propose a deep learning...

a new scheme was designed to recognize the medicine which was packed in small cylindrical bottle with a label on the surface. To recognize the label of medicine a new algorithm was needed, which was used to rectify the cylindrical character to plane one. According to the equations of the rectification...

Visual tracking is an important research direction in the field of computer vision and has been widely used in military, medical and other fields. In recent years, the upsurge of deep learning in computer vision provides a new way for the realization of visual tracking with higher performance. This paper...

Incremental learning is a machine learning paradigm that the training dataset grows dynamically at runtime and the prediction accuracy can be improved incrementally. Since real-life datasets are usually open-ended and dynamic, incremental learning is important to many image classification applications....

In this paper, we use one fast Fourier algorithm of the Fresnel integral transform (S-FFT) to realize the digital watermarking information of hologram. The most important point is that by the four-step phase shifting method to obtain four holographic images. We can randomly choose a holographic image...

We proposed a novel method of people recognition based on the structure of bones, which is detected by Kinect sensor for skeletal tracking. By using 3D depth image achieved by Kinect, body's skeleton information is extracted as 20 joints with joint coordinates, from which body's skeletal features are...

The design and development of complex 2D symbol library system based on GIS system is completed by combining design pattern technology. Firstly, it states that the design of the symbol library system has its unique difficulties. The diversity and complexity of symbol library lead to difficulties in the...

In recent years, air pollution problem has been the focus of public attention. In this paper, we proposed an efficient algorithm to evaluate the Air Quality Index (AQI) based on image recognition technology. In offline stage, some distinctive features extracted from the photos which are captured by common...

This paper introduces a method that can detect human posture by using depth image. The method uses head model to locate the human position which includes edge extraction, template matching and human detection. Then we extract the HOG feature from the depth images to get the characteristic vector of the...

How to improve the quality of difference image (DI) for change detection task is an important issue in remote sensing images. This paper propose a new DI creation method based on deep neural networks. Deep belief network (DBN) which is an important model in deep learning is applied, and back propagation...

Manual ways of recognizing stored grain pests which are trapped is very time-consuming. In this study, an image dataset of 9 species of pests was built up by finding the MSERs (Maximally Stable Extremal Regions), through a trap of stored grain pests combined with a real-time imaging device. On this basis,...

The complicate calculation and frequent comparison based on spatial orientation tree (SOT) structure in SPIHT severely reduce the coding speed. To overcome these shortcomings, a modified image compression algorithm is proposed without SOT. The algorithm sorts the wavelet coefficients in descending order...

With the rapid development of computer technology and the increasing of multimedia data on the Internet, how to quickly find the desired information in the massive data becomes a hot issue. Image retrieval can be used to retrieve similar images, and the effect of image retrieval depends on the selection...

When recognizing facial expression sequences by discrete Hidden Markov Model, it is necessary to cluster the image frames into several observation states. Considering the complexity of face images in emotion expressing, we partition the whole face into several sub-regions to do clustering separately....

With the development of the tourism, tourism virtual visualization became the essential method to show the charm of scenic spots. Study on the integration technology of panoramic images, 3D GIS and cloud GIS to visualize the mountain scenic spots in order to meet the needs of travelers, the integration...