Aiming at the problem of nearshore ship detection in highresolution optical remote sensing images, this paper proposed a port ship target detection method based on a lightweight multi-scale convolutional network. After verification, the method has a good detection effect on port ships. The network improves the feature expression ability without increasing the computational complexity, and can effectively capture rotation-sensitive features, thereby improving the versatility of rotating samples. The average detection rate of all types of ships in the experiment is 96.92%, and the average false alarm rate is 8.54%. High detection rate of ship target can be guaranteed and various false alarm target interference can be effectively eliminated.
The quality of remote sensing images is directly influenced by the quantization bits of the CCD camera in the satellite. The images with higher quantization bits can contain more feature details of ground objects, therefore, are considered to be of higher quality. But high quantization bits also mean large amounts of data, and the images acquired by the CCD camera need to be compressed to shorten the satellite-to-ground transmission time. The existed compressed image quality evaluation model can hardly assess the loss degree of images with high quantization bits effectively. This article briefly described the main process of evaluation of compressed remote sensing images. And a new evaluation method with both subjective and objective factors was set up for the compressed images with high quantization bits and tested with actual satellite remote sensing images. The results show that the proposed evaluation method is consistent with the human eye subjective feeling and can provide quantitative reference for the comprehensive evaluation of the remote sensing satellite.
Microscope is the primary scientific instrument in many laboratories. Nowadays, with the development of science and technology, requirements on the performance of microscopic imaging are growing rapidly. Light field microscopy (LFM) is an effective approach of obtaining three-dimensional (3D) information. However, the LFM compromises the spatial resolution of image. To solve this problem, this paper proposes a new method by combining LFM with Fourier ptychographic (FP) algorithm, which iteratively stitches together a number of variably illuminated, low-resolution intensity images in Fourier space to produce a wide-field, high-resolution complex sample image. The hardware implement of the system is mainly introduced, which contains the image system and the illumination system. This system uses epi-illumination for non-transparent sample image. To verify the capability of this system, experiments have been done. Firstly, a 150 μm size micro-lens array was used to image without FP algorithm. Secondly, FP algorithm was added to the experiments. Preliminary results showed the potential of the 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.