Object detection technology has always been one of the important research directions in the field of computer vision. It has important application value and prospects in both civil fields and military fields. With the emergence of artificial intelligence technology, deep learning has gradually replaced the traditional algorithm with its higher accuracy. Considering that most of the current algorithms are for color image, compared with color image, infrared image contains less feature information, and it is more difficult in object detection. In this paper, different algorithms are used for object detection in infrared images, and the detection results are compared. This paper chooses YOLO V5 and combines it with MobileNet to lighten the model. After lightening, the parameters are reduced by 30%, but the accuracy is only reduced by 5%. Finally, this paper quantify YOLO V5 based on the model quantization method of PyTorch. After quantization, the accuracy of the model decreases by 2%.
In the polarization imaging of shortwave infrared sea surface background, the sea surface flare as a strong interference source seriously reduces the contrast of the image, resulting in low probability of target detection, high false alarm rate, easy to be lost in tracking and other problems, which must be suppressed. The traditional solar flare suppression method suppresses the s component of the solar flare by adding a horizontal polarization filter, ignoring the p component, resulting in a general effect on flare suppression, and can not establish a relationship with the position of the sun and the position of the detector. real-time suppression of solar flare. To solve this problem, a real-time method to suppress solar flare is proposed in this paper. Combined with Cox-Munk model and polarization bi-directional reflection distribution function (pBRDF), the rough sea surface flare reflection model is established, and the polarization filter is adjusted according to the polarization angle of the flare. The related polarization degree and polarization angle images are simulated by MATLAB, and the results are in good agreement with the reality, which verifies the accuracy of the model.
Infrared weak and small target detection has important application value in military field, and is a hot research topic in the field of target detection. With the research and development of technology, some guiding and innovative detection algorithms are emerging, especially the advantage of machine learning algorithm. In this paper, the infrared week and small target detection is considered as the two classification problem of target and background, and an infrared week and small target detection algorithm based on multiple features SVM posterior probability is proposed and applied to the weekly vision search system. In the experiment, the SVM classification model is obtained by using 8 good segmentation characteristics as the main reference value of the identification target and background, and training the training set through a large number of target and background samples. Finally, the SVM posterior probability is selected as the output of the detection result.The simulation experiment of this algorithm and the application experiment in the transplant week search system show that the method of this method has higher target detection probability (not less than 95%),the false alarm probability is lower, and the time complexity of the algorithm is low, and the hardware resource is less, and it has a good prospect of engineering application.
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.