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 5. Отображено 5.
06-06-2023 дата публикации

Video saliency prediction method based on double-U structure

Номер: CN116229106A
Автор: YAN CHENGGANG, WU SONGHE, CHEN YUZHONG, ZHOU XIAOFEI, GAO YUHAN, SUN YAOQI, ZHU ZUNJIE, CHEN CHUQIAO, WANG HONGKUI, WANG TINGYU, YIN HAIBING, ZHANG JIYONG, LI ZONGPENG, ZHAO ZHIDONG
Принадлежит:

The invention discloses a video saliency prediction method based on a double-U structure, and the method comprises the steps: effectively mining global semantic information and local detail information through employing a novel double-U coding and decoding structure, and enabling the more full utilization of multi-scale features; by using a bridging block, the receptive field of the model is further expanded, and the capability of acquiring multi-scale context information of the model is enhanced; by deploying a weighted fusion module at each level of an encoder and a decoder, temporal and spatial features are fused more effectively. Through the design, the performance of video saliency prediction can be further improved.

Подробнее
Номер записи: 1
15-12-2010 дата публикации

Accumulator charging and discharging control method for smoothening power fluctuation of wind power station

Номер: CN0101917014A
Автор: WENXIA PAN, ZHONGXING FU, JIANQIANG CHEN, PENGFEI WANG, XIANG WANG, CHENGXIANG SUN, HAIPING HE, SONGHE WU, HUAN GUO, PAN WENXIA, FU ZHONGXING, CHEN JIANQIANG, WANG PENGFEI, WANG XIANG, SUN CHENGXIANG, HE HAIPING, WU SONGHE, GUO HUAN
Принадлежит:

The invention provides an accumulator charging and discharging control method for smoothening the power fluctuation of a wind power station, belonging to the technical field of wind power generation control. The control method comprises the following steps of: charging the accumulator when the capacity of the accumulator satisfies the operation requirement and the active power is larger than a permissible upper limit value of a power grid; and discharging the accumulator when the active power is smaller than a permissible lower limit value of the power grid, or else enabling the accumulator to exit the operation. The invention estimates the residual capacity of the accumulator by detecting the active power output by the system and the charge state of the accumulator to control the operation mode of the accumulator and achieve the aim of smoothening the output power fluctuation of the wind power station; and the reactive loss of a converter is reduced by adopting a PWM (Pulse Width Modulation ...

Подробнее
Номер записи: 2
09-05-2023 дата публикации

Road surface crack detection method based on coder-decoder

Номер: CN116091426A
Автор: YAN CHENGGANG, CHEN YUZHONG, YANG HAONAN, ZHANG WENHAO, WU SONGHE, ZHU ZUNJIE, GAO YUHAN, SUN YAOQI, CHEN CHUQIAO, WANG HONGKUI, WANG TINGYU, YIN HAIBING, ZHANG JIYONG, LI ZONGPENG, ZHAO ZHIDONG
Принадлежит:

The invention discloses a pavement crack detection method based on a coder-decoder, which can effectively detect the crack of a pillow layer by constructing a new end-to-end coder-decoder residual network. By combining a deep supervision mechanism and fusion loss, the model can capture fine details, so that the network is easier to optimize. The model can directly output a high-quality saliency map which is almost close to a corresponding ground true value. The obtained saliency mapping uniformly highlights defect objects with clear boundaries, and effectively filters background noise at the same time. The model provided by the invention has relatively high robustness, does not need any post-processing, and is relatively high in real-time speed on a single GPU (Graphics Processing Unit).

Подробнее
Номер записи: 3
13-06-2012 дата публикации

Accumulator charging and discharging control method for smoothening power fluctuation of wind power station

Номер: CN0101917014B
Автор: PAN WENXIA, FU ZHONGXING, CHEN JIANQIANG, WANG PENGFEI, WANG XIANG, SUN CHENGXIANG, HE HAIPING, WU SONGHE, GUO HUAN, WENXIA PAN, ZHONGXING FU, JIANQIANG CHEN, PENGFEI WANG, XIANG WANG, CHENGXIANG SUN, HAIPING HE, SONGHE WU, HUAN GUO
Принадлежит:

The invention provides an accumulator charging and discharging control method for smoothening the power fluctuation of a wind power station, belonging to the technical field of wind power generation control. The control method comprises the following steps of: charging the accumulator when the capacity of the accumulator satisfies the operation requirement and the active power is larger than a permissible upper limit value of a power grid; and discharging the accumulator when the active power is smaller than a permissible lower limit value of the power grid, or else enabling the accumulator to exit the operation. The invention estimates the residual capacity of the accumulator by detecting the active power output by the system and the charge state of the accumulator to control the operation mode of the accumulator and achieve the aim of smoothening the output power fluctuation of the wind power station; and the reactive loss of a converter is reduced by adopting a PWM (Pulse Width Modulation ...

Подробнее
Номер записи: 4
23-06-2023 дата публикации

Video saliency prediction method based on gated fusion network

Номер: CN116310695A
Автор: YAN CHENGGANG, WU SONGHE, ZHOU XIAOFEI, SUN YAOQI, ZHU ZUNJIE, WANG SHUAI, GAO YUHAN, WANG HONGKUI, YIN HAIBING, ZHANG JIYONG, LI ZONGPENG, ZHAO ZHIDONG
Принадлежит:

The invention discloses a video saliency prediction method based on a gating fusion network. The method comprises the following steps: firstly, constructing the gating fusion network; the method comprises the following steps: extracting spatio-temporal features contained in a video clip through an encoder, enhancing top coding features by using a bridging module, fusing coding and decoding features by using a gating fusion module, and finally decoding the gating fusion features to predict a final saliency map. The invention provides a gating fusion module which is used for effectively fusing the multi-level characteristics of the encoder into the decoder. The module weights and encodes features along time, space and channel dimensions, so that the position of a significant object is more accurately positioned, and redundant spatio-temporal information in the features is filtered out.

Подробнее
Номер записи: 5