This paper presents a convolutional neural networks (CNN) based on sparse coding for human postures recognition. It’s an unsupervised approach for color multi-channel processing. The improvement of the method is mainly reflected in two aspects. We transform sample images into patches and make a decorrelation between input patches and reconstructed patches. In addition, we use the convolution kernels extracted by sparse coding to replace the initialization of the convolution kernels for human postures recognition. The proposed method is tested in the public KTH pedestrian behavior dataset and HUMAN-V2 self-test dataset. Compared with the traditional way, our approach shortens the training time a lot and also improves the recognition rate. Our experimental results verifies the effectiveness.
This paper presents a new non-uniformity correction algorithm for Infrared Focal Plane Arrays. After some
background on the research of non-uniformity correction, we highlight the adaptive scene-based techniques and
algorithm proposed which attempt to realize the correction. In our algorithm, the compensation coefficients are
adaptively adjusted with the artificial neural networks. After this algorithm was derived, we also analyzed the
performance of algorithm under the real conditions. The result shows that our algorithm can effectly reduce the
undesired fixed-pattern noise and compensated the time-vary coefficients.
Recognition of the interesting targets is the key techniques of precise guided weapon systems. Because fractal dimension is an interesting textual feature of an image, it has been used in many pattern recognition applications including classification and segmentation. According to the fractal feature of man-made objects in infrared images, a new algorithm is presented to detect the airplanes in this paper. And then we can partition and identify the potential targets using this fractal algorithm. Simulations illustrate that the airplane is successfully identified with the algorithm. The algorithm only requires moderate operations, so it is easy to be implemented for automatic target detection in real-time systems. The results of the experiments show that the fractal dimension can efficiently reflect the object surface complexity or irregularity in images. The algorithm is a powerful tool in identifying airplanes from infrared images.
KEYWORDS: Target detection, Infrared imaging, Digital signal processing, Wavelet transforms, Infrared radiation, Image processing, Field programmable gate arrays, Wavelets, Detection and tracking algorithms, Linear filtering
The algorithm and the architecture of the information processing system of infrared imaging guidance are studied. Wavelet transform can decompose the image in multi-scale, so the different frequencies components of an image can be selected. The dim target can be detected by wavelet transform. The dramatic increase in the computational speed provided by DSP offers the possibilities of processing large high-resolution images in real time. The authors have experience in processing high-resolution images (up to pixels, eight bits per pixel) using TI's TMS320C6201 DSP. We have developed a technique, which uses FPGA to perform wavelet transform to decrease the execution time.
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.