This article proposes a Facial Expression Hierarchical Detection Network (FHDN), use a convolutional neural network based on multiple branches, about facial expression detection. To further improve feature extraction performance, the method adds an ESSAM module as an attention mechanism. The ESSAM module can adaptively adjust the weight of each feature map, thus improving the performance of the model in feature extraction and facial expression recognition tasks. The method was experimentally evaluated on a self-made dataset, and the results showed that the detection accuracy rate of this model was 81.40%, which is an improvement of 5% and 0.7% compared to YOLOV5 and YOLOV8, respectively. When compared to conventional deep learning techniques, this approach extracts picture characteristics more quickly and accurately without the need for labor-intensive manual labor.
Due to inclement weather caused frequently, such as clouds, fog , rain etc. The light intensity on the illuminated objects falls sharply, it make the scenes captured unclear, poor visual quality and low contrast degree. To improve the overall quality of these images, especially the bad illuminated images, the paper propose a new color image enhancement algorithm which is based on multi-scale Retinex theory with color recovering factor (MSRCR) and the human visual system (HVS). It can effectively solve the problem of the color balance of digital images by removing the influence of light and obtain component images reflected the reflex of the object surface, meanwhile, reduce the impact of non-artificial factors and overcome the Ringing effect and human interference. Through comparison and contrast among experiments, that combined evaluated parameters on enhancement image, such as variance, average gradient, sharpness and so forth with the traditional image enhancement methods, such as histogram enhancement, adaptive histogram enhancement, the MSRCR algorithm is proved to be effective in image contrast, detail enhancement and color fidelity, etc.
KEYWORDS: CMOS sensors, Digital signal processing, Image processing, Imaging systems, Image acquisition, Control systems, Image interpolation, Image quality, Data storage, Data acquisition
An image acquisition system is introduced, which consists of a color CMOS image sensor (OV9620), SRAM
(CY62148), CPLD (EPM7128AE) and DSP (TMS320VC5509A). The CPLD implements the logic and timing control to
the system. SRAM stores the image data, and DSP controls the image acquisition system through the SCCB (Omni
Vision Serial Camera Control Bus). The timing sequence of the CMOS image sensor OV9620 is analyzed. The imaging
part and the high speed image data memory unit are designed. The hardware and software design of the image
acquisition and processing system is given. CMOS digital cameras use color filter arrays to sample different spectral
components, such as red, green, and blue. At the location of each pixel only one color sample is taken, and the other
colors must be interpolated from neighboring samples. We use the edge-oriented adaptive interpolation algorithm for the
edge pixels and bilinear interpolation algorithm for the non-edge pixels to improve the visual quality of the interpolated
images. This method can get high processing speed, decrease the computational complexity, and effectively preserve the
image edges.
KEYWORDS: CMOS sensors, Image acquisition, Image segmentation, Digital signal processing, Control systems, Imaging systems, Image processing, Data storage, Cameras, Optical character recognition
An image acquisition system is introduced, which consists of a color CMOS image sensor (OV9620), SRAM
(CY62148), CPLD (EPM7128AE) and DSP (TMS320VC5509A). The CPLD implements the logic and timing control to
the system. SRAM stores the image data, and DSP controls the image acquisition system through the SCCB (Omni
Vision Serial Camera Control Bus). The timing sequence of the CMOS image sensor OV9620 is analyzed. The imaging
part and the high speed image data memory unit are designed. The system structure and its application of CMOS image
sensor OV9620 in paper currency number recognition are also introduced. The hardware and software design of the
image acquisition and recognition system is given. In this system, we use the template matching character recognition
method to guarantee fast recognition speed and high correct recognition probability.
The design of the high mass image storage system is introduced using DSP, FPGA and Flash structure. Texas
Instruments Corporation DSP chip (TMS320VC5509APEG) is used as the main controller, Samsung's Flash chips
(K9F2G08U0M) used as the main storage medium, and the Xilinx Corporation FPGA chip (XCV600E) used as logic
control modules. In this system, Storage module consists of 32 Flash memory chips, which are divided into 8 groups that
correspond to 8-level pipeline. The 4-Flash memory chip forms a basic 32-bit memory module. The entire system
storage space is 64 G bit. Through simulation and verification, the storage speed is up to 352Mbps and readout speed is
up to 290Mbps, it can meet the demand to the high-speed access, and which has strong environmental adaptability.
KEYWORDS: Stars, Electron multiplying charge coupled devices, Signal to noise ratio, Charge-coupled devices, Interference (communication), Photons, Signal detection, Signal processing, Image quality, Optical tracking
Star tracker is a high precise, high reliable attitude measurement component of spacecrafts, which plays
a very important role in attitude measurement and control system. Unfortunately, in low light level, in
order to obtain good quality image, the common CCD needs more time to integral light. This leads low
data output and update slowly. In this paper, the star tracker based on a novel electronic multiply CCD
(EMCCD) is introduced. EMCCD has very high sensitivity. The application of EMCCD is very finite
in space explore field, although it has been applied by some country. At first, the detection sensitivity
of EMCCD is analyzed, based on the signal detection theory in noise and optimal SNR threshold
detection principle, and the detection sensitivity model is established. And then, the main noise sources
of EMCCD are analyzed. Finally, as an example, a specific detection sensitivity calculation of EMCCD
star tracker is provided with given optical parameters and exposure time.
Star tracker is the most precision attitude measurement instrument of spacecrafts, which plays a very important role in
attitude measurement and control system. The technology of star tracker based on CCD is very popular. Unfortunately,
with a CCD system, a single integration period for the entire sensor is necessitated. This leads low data output and
update slowly. Moreover, CCD star tracker is not proper to micro-spacecraft because its volume, weight and energy
consume cannot further decrease. Thus, the novel electron-multiplying CCD (EMCCD) is starting their way in space
applications field. Centroiding algorithm is a subpixel position determination method proper to star position calculation
because of its high accuracy and simplicity. But its position accuracy is affected by various kinds of noise. In this paper,
the subpixel position accuracy of centroiding algorithm is analyzed. The focus is on the estimation of the attainable
accuracy in the application of EMCCD detector, and analysis of the EMCCD noise influence on the star subpixel
position accuracy to find the methods of improving it. The analysis shows that the main contributors to the errors of
subpixel position come from dark noise, read noise and photon shot noise. Simulation experiment results show that the
subpixel position accuracy can attain 1/45 pixel, which
Recent research results show that the low-frequency noise of optoelectronic coupled devices has become an important
sensitive parameter affecting its operational status and reliability. Screening optoelectronic coupled devices is an
effective method by measuring the noise power spectrum. However, this method is based on the traditional Fourier
analysis and spectrum analysis, its measuring speed is quite slow, and the method used to establish screening threshold is
more complicated. In this paper, a set of measurement and analysis system based on virtual instrument is set up, which is
composed of dual-channel low-noise pre-amplifiers, dynamic signal analyzer Agilent35670A and PC. According to the
wavelet analysis method, the different kinds of noise can be identified. Through the GPIB control, separating the 1/f
noise, the g-r noise and the burst noise is performed and the noise analysis process is finished by the LabVIEW
procedure. Experimental results demonstrate that this system can not only improve reliability of screening device to
satisfy higher reliability and quality requirement, but also the testing and analyzing process is finished faster and more
accurately than the traditional 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.