Настройки

Укажите год
-

Небесная энциклопедия

Космические корабли и станции, автоматические КА и методы их проектирования, бортовые комплексы управления, системы и средства жизнеобеспечения, особенности технологии производства ракетно-космических систем

Подробнее
-

Мониторинг СМИ

Мониторинг СМИ и социальных сетей. Сканирование интернета, новостных сайтов, специализированных контентных площадок на базе мессенджеров. Гибкие настройки фильтров и первоначальных источников.

Подробнее

Форма поиска

Поддерживает ввод нескольких поисковых фраз (по одной на строку). При поиске обеспечивает поддержку морфологии русского и английского языка
Ведите корректный номера.
Ведите корректный номера.
Ведите корректный номера.
Ведите корректный номера.
Укажите год
Укажите год
Применить Всего найдено 7. Отображено 7.
04-04-2023 дата публикации

Underwater image contrast enhancement method based on double prior optimization

Номер: CN115908179A
Автор: ZHANG WEIDONG, LI GUOHOU, JIN SONGLIN, ZHOU LING, LI TENGFEI, QU PEIXIN, GAO GUOHONG, WANG JIANPING, AN JINLIANG, ZHAO GAOLI
Принадлежит:

The invention provides an underwater image contrast enhancement method based on double prior optimization, and the method comprises the steps: obtaining an underwater image after color correction, converting the underwater image after color correction from an RGB color model to an HSV color model, and carrying out the decomposition of a basic layer and a detail layer for a V channel, spatial prior and texture prior are adopted to decompose a V channel into a base layer and a detail layer, an integral strategy is utilized to count local mean and variance for the base layer to realize local contrast enhancement of the base layer, and a nonlinear stretching function is utilized to realize texture detail enhancement for the detail layer. According to the double prior optimization underwater image contrast enhancement method, the contrast of the image is enhanced, texture details are highlighted, and the problem of texture detail loss caused by scattering can be well solved.

Подробнее
Номер записи: 1
15-07-2015 дата публикации

Multi-phase-jump and vehicle full-dynamic induction traffic control method

Номер: CN104778845A
Автор: BAI LINFENG, QU PEIXIN, WANG YINGJUN, ZHAO MINGFU
Принадлежит:

The invention relates to an urban intersection traffic signal control method, relates to a multi-phase-jump and vehicle full-dynamic induction traffic control method, and aims to flexibly control traffic signals at intersections by using a traffic signal instrument so as to improve the traffic pass property of vehicles at the intersections and reduce the total delay time of vehicles. The method comprises the following steps: 1) setting an upstream vehicle detector and a parking line vehicle detector at each lane of a traffic intersection, and detecting dissipating vehicle and information of vehicles entering the intersection in real time; 2) according to the numbers of vehicles waiting in different lanes and vehicles running into the intersections, automatically transmitting passing signals to busy lanes by using a busy lane priority pass method through a traffic signal instrument control system so as to induce the vehicles in the lanes to pass prior; 3) by using a method that compatible ...

Подробнее
Номер записи: 2
18-07-2023 дата публикации

Near infrared spectrum medicine identification method based on combined strategy

Номер: CN116453610A
Автор: ZHANG WEIDONG, ZHENG YING, ZHOU LING, LI GUOHOU, LI ZEXU, JIN SONGLIN, TIAN YONGQIN, ZHAO WENYI, QU PEIXIN
Принадлежит:

The invention provides a near infrared spectrum drug identification method based on a combined strategy, which comprises the following steps of: firstly, constructing the advantage of a deep convolutional network by utilizing a residual mechanism to improve the feature extraction capability of a spectrum, and then, carrying out model transfer on small sample data by utilizing the advantage of overcoming the data sensitivity of transfer learning to obtain the drug identification result. According to the method, the model lifting rate is higher and the convergence effect is better in the training process, and finally, a Focal Loss loss function is proposed for the imbalance of a source data set, so that a sample with a smaller expected scale can also make a greater contribution to the loss, and the tendency of the model to most samples is solved. According to the method, after the network depth is deepened, the drug spectrum feature extraction capacity is higher, model cross-drug type transfer ...

Подробнее
Номер записи: 3
23-05-2023 дата публикации

Wheat seed classification method

Номер: CN116152524A
Автор: AN JINLIANG, ZHANG CHEN, LI YAQIANG, JIN SONGLIN, ZHENG YING, ZHANG WEIDONG, LI GUOHOU, QU PEIXIN, WANG YINGJUN, ZHANG ZIYANG
Принадлежит:

The invention provides a wheat seed classification method, which is a method based on an ICNMF fusion method and a 3DResnetCBAM network, and mainly comprises the following steps: respectively carrying out background separation on a hyperspectral image and an RGB image of a wheat seed, respectively obtaining an RGB reflectivity image and a low-spatial-resolution hyperspectral image, and carrying out image segmentation on the RGB reflectivity image and the low-spatial-resolution hyperspectral image; and the seeds are fused into hyperspectral images with high spatial resolution and high spectral resolution by using an ICNMF fusion method, and then the hyperspectral images are sent to the constructed 3DResnetCBAM network for identification to obtain seed varieties. According to the method, the RGB image which is easy to obtain is adopted to enhance the hyperspectral image features, the collection cost is reduced, and then the cost of wheat seed classification and recognition is reduced.

Подробнее
Номер записи: 4
27-06-2023 дата публикации

Wheat seed classification method based on hyperspectral low-rank tensor decomposition

Номер: CN116342933A
Автор: AN JINLIANG, ZHENG YING, JIN SONGLIN, ZHANG CHEN, LI YAQIANG, ZHANG WEIDONG, LI GUOHOU, QU PEIXIN, WANG YINGJUN, ZHANG ZIYANG
Принадлежит:

The invention provides a wheat seed classification method based on low-rank tensor decomposition. The wheat seed classification method comprises the steps of inputting a hyperspectral image, normalizing gray values of all pixels to be 0-1, separating a wheat seed region from a background region by adopting a threshold segmentation method, obtaining a background mask matrix, and calculating the hyperspectral average reflectivity of each seed by using the obtained mask. The method comprises the following steps: expressing original data as a tensor form, and performing dimension reduction on an original tensor by using a low-rank tensor decomposition method to obtain a tensor after dimension reduction; and inputting the generated dimensionality-reduced data into a combined classifier for classification, grouping classification results according to the seeds to which the classification results belong, and selecting the most classification results in each group of results as the group, namely ...

Подробнее
Номер записи: 5
04-04-2023 дата публикации

Underwater image classification depth cascade network based on unified multi-color model learning

Номер: CN115908919A
Автор: ZHANG WEIDONG, ZHOU LING, LI GUOHOU, LI ZEXU, JIN SONGLIN, ZHENG YING, QU PEIXIN, GAO GUOHONG, ZHAO GAOLI, WANG JIANPING
Принадлежит:

The invention provides an underwater image classification depth cascade network based on unified multi-color model learning, and the method comprises the steps: firstly providing a multi-color model feature coding strategy, and integrating the features of different color models into a unified feature model, so as to enrich the diversity of feature representation capability; a deep support vector machine is designed by considering that a deep network structure has better nonlinear modeling performance; meanwhile, a unified multicolor model is embedded into the deep cascade network, and a gradient descent reverse optimization method is adopted to optimize the network model. The method solves the problems that when a traditional fusion and classification method is used for classifying underwater images with degraded colors, more feature quantities and large parameters are easily introduced, and the problems that a single-color model is insufficient in learning feature, low in classification ...

Подробнее
Номер записи: 6
18-07-2023 дата публикации

Self-supervised visual representation learning method based on semantic and position information

Номер: CN116451574A
Автор: ZHANG WEIDONG, ZHOU LING, JIN SONGLIN, ZHENG YING, TIAN ZHEN, LI GUOHOU, TIAN YONGQIN, QU PEIXIN, ZHAO WENYI
Принадлежит:

The invention provides a self-supervised visual representation learning method based on semantic and position information, and the method comprises the following steps: building a model which comprises a sampling module, a feature extractor and a cascade collaborative optimization module; a semantic and position constraint rule is added into the sampling module, and the sampling module is used for processing an original image and extracting data containing semantic and position constraints; the feature extractor is used for generating vectors, processing extracted data containing semantic and position constraints and sending the processed data to the cascade collaborative optimization module; the cascade collaborative optimization module comprises two tasks, namely a decoupling task and a comparison task, and the two tasks adopt a cascade normal form to capture complementary characteristics of a semantic relation and a position relation. According to the method, a corresponding decoupling ...

Подробнее
Номер записи: 7