Deep learning-based object detection networks outperform the traditional detection methods. However, they lack interpretability and solid theoretical guidance. To guide and support the application of object detection networks in infrared images, this work analyzes the influence of infrared image quantization on the performance of object detection networks. Firstly, the traditional infrared quantization methods and deep learning-based object detection networks are introduced, and the characteristics of these methods are analyzed. Then, the influence of four typical quantization methods on the performances of two object detection networks is compared, and the influence mechanism is analyzed through a cross-comparison experiment. The experimental results show that infrared image quantization is more helpful for learning the discriminative feature of the object/background for the object detection networks. Moreover, the feature difference of object/background caused by different quantization methods will seriously affect the performance of object detection networks. The research provides support for the application of deep learning-based detection networks in infrared scenes, which is of great significance.
The first part of the paper briefly introduces the automatic speech recognition (ASR) technology, mainly about its development status and its implementation steps. The second part summarizes the current technical bottlenecks and related solutions in this field, and also predicts the prospects.
A real-time multi-channel high-definition image mosaic system is designed based on FPGA. First, the best matching pointsare obtained by the improved SIFT feature point matching algorithm. Wide Angle camera's internal and external parameters are calculated using the image calibration method. According to sphere reverse projection principle and the camera’s parameters, transform matrix are generated. Then, FPGA can completetransformation,projection,mosaic and display for each high-definition image in parallel.In this paper, thetransform matrix is calculated by image calibration algorithm.Multi-channel high-definition video mosaic system based on FPGA is realized. The design has very strong practicability.
KEYWORDS: Visualization, Optoelectronics, 3D modeling, 3D image processing, Data modeling, Data fusion, Associative arrays, Visual process modeling, Data communications, Image processing
Introduces the image and spatial data fusion in photoelectric detection application. Focus on the optimization of 3D scene reconstruction. In airborne imaging applications, aiming at the problems of massive terrain data, this paper proposed a dynamic data scheduling strategy which is based on state-tree from simplification, and present a terrain data dynamic schedule framework from render optimization. For the suggested optimized procedure and framwork, give a experiment and couclusion based on programmimg, it prove that the suggested dynamic schedule strategy in this paper could fastly construct three-dimensional scene in flight simulation, could speed up the three-dimensional visulization, it could meet the practical requiremnet of engineering in flight simulation.
In this paper, we propose a high definition video display format conversion system based on FPGA which converts the Camera Link format video data (1928×1084, progressive scanning, 33 fps) into the HDMI format video (1920×1080p, 30 fps), and then output to the HD display device with a HDMI interface. This system can solve the display incompatible problem between different HD video frame rates. Experimental results show that the proposed system realizes a real-time video format conversion; furthermore, the output video image has a good quality.
In order to solve the problem of the bandwidth limitation of the image transmission system on UAV, a scheme with image compression technology for mini UAV is proposed, based on the requirements of High-definition image transmission system of UAV. The video codec standard H.264 coding module and key technology was analyzed and studied for UAV area video communication. Based on the research of high-resolution image encoding and decoding technique and wireless transmit method, The high-resolution image transmission system was designed on architecture of Android and video codec chip; the constructed system was confirmed by experimentation in laboratory, the bit-rate could be controlled easily, QoS is stable, the low latency could meets most applied requirement not only for military use but also for industrial applications.
For unmanned aerial vehicle(UAV) images, the sensor can not get high quality images due to fog and haze weather. To solve this problem, An improved dehazing algorithm of aerial high-definition image is proposed. Based on the model of dark channel prior, the new algorithm firstly extracts the edges from crude estimated transmission map and expands the extracted edges. Then according to the expended edges, the algorithm sets a threshold value to divide the crude estimated transmission map into different areas and makes different guided filter on the different areas compute the optimized transmission map. The experimental results demonstrate that the performance of the proposed algorithm is substantially the same as the one based on dark channel prior and guided filter. The average computation time of the new algorithm is around 40% of the one as well as the detection ability of UAV image is improved effectively in fog and haze weather.
The electronic image stabilization technology based on improved optical-flow motion vector estimation technique can effectively improve the non normal shift, such as jitter, rotation and so on. Firstly, the ORB features are extracted from the image, a set of regions are built on these features; Secondly, the optical-flow vector is computed in the feature regions, in order to reduce the computational complexity, the multi resolution strategy of Pyramid is used to calculate the motion vector of the frame; Finally, qualitative and quantitative analysis of the effect of the algorithm is carried out. The results show that the proposed algorithm has better stability compared with image stabilization based on the traditional optical-flow motion vector estimation method.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.