Natural scene recognition has important significance and value in the fields of image retrieval,autonomous navigation,human-computer interaction and industrial automation.Firstly,the natural scene image non-text conte...Natural scene recognition has important significance and value in the fields of image retrieval,autonomous navigation,human-computer interaction and industrial automation.Firstly,the natural scene image non-text content takes up relatively high proportion;secondly,the natural scene images have a cluttered background and complex lighting conditions,angle,font and color.Therefore,how to extract text extreme regions efficiently from complex and varied natural scene images plays an important role in natural scene image text recognition.In this paper,a Text extremum region Extraction algorithm based on Joint-Channels(TEJC)is proposed.On the one hand,it can solve the problem that the maximum stable extremum region(MSER)algorithm is only suitable for gray images and difficult to process color images.On the other hand,it solves the problem that the MSER algorithm has high complexity and low accuracy when extracting the most stable extreme region.In this paper,the proposed algorithm is tested and evaluated on the ICDAR data set.The experimental results show that the method has superiority.展开更多
Visual question answering(VQA)has attracted more and more attention in computer vision and natural language processing.Scholars are committed to studying how to better integrate image features and text features to ach...Visual question answering(VQA)has attracted more and more attention in computer vision and natural language processing.Scholars are committed to studying how to better integrate image features and text features to achieve better results in VQA tasks.Analysis of all features may cause information redundancy and heavy computational burden.Attention mechanism is a wise way to solve this problem.However,using single attention mechanism may cause incomplete concern of features.This paper improves the attention mechanism method and proposes a hybrid attention mechanism that combines the spatial attention mechanism method and the channel attention mechanism method.In the case that the attention mechanism will cause the loss of the original features,a small portion of image features were added as compensation.For the attention mechanism of text features,a selfattention mechanism was introduced,and the internal structural features of sentences were strengthened to improve the overall model.The results show that attention mechanism and feature compensation add 6.1%accuracy to multimodal low-rank bilinear pooling network.展开更多
The robust guarantee of train control on-board equipment is inextricably linked to the safe functioning of a high-speed train.A fault diagnostic model of on-board equipment is built utilizing the integrated learning X...The robust guarantee of train control on-board equipment is inextricably linked to the safe functioning of a high-speed train.A fault diagnostic model of on-board equipment is built utilizing the integrated learning XGBoost(eXtreme Gradient Boosting)algorithm to help technicians assess the malfunction category of high-speed train control on-board equipment accurately and rapidly.The XGBoost algorithm iterates multiple decision tree models to improve the accuracy of fault diagnosis by lifting the predicted residual and adding regular terms.To begin,the text features were extracted using the improved TF-IDF(Term Frequency-Inverse Document Frequency)approach,and 24 fault feature words were chosen and converted into weight word vectors.Secondly,considering the imbalanced fault categories in the data set,the ADASYN(Adaptive Synthetic sampling)adaptive synthetically oversampling technique was used to synthesize a few category fault samples.Finally,the data samples were split into training and test sets based on the fault text data of CTCS-3train control on-board equipment recorded by Guangzhou Railway Group maintenance personnel.The XGBoost model was utilized to realize the automatic fault location of the test set after optimized parameter tuning through grid search.Compared with other methods,the evaluation index of the XGBoost model was significantly improved.The diagnostic accuracy reached 95.43%,which verifies the effectiveness of the method in text fault diagnosis.展开更多
基金This work is supported by State Grid Shandong Electric Power Company Science and Technology Project Funding under Grant Nos.520613180002,62061318C002the Fundamental Research Funds for the Central Universities(Grant No.HIT.NSRIF.201714)+1 种基金Weihai Science and Technology Development Program(2016DX GJMS15)Key Research and Development Program in Shandong Provincial(2017GGX90103).
文摘Natural scene recognition has important significance and value in the fields of image retrieval,autonomous navigation,human-computer interaction and industrial automation.Firstly,the natural scene image non-text content takes up relatively high proportion;secondly,the natural scene images have a cluttered background and complex lighting conditions,angle,font and color.Therefore,how to extract text extreme regions efficiently from complex and varied natural scene images plays an important role in natural scene image text recognition.In this paper,a Text extremum region Extraction algorithm based on Joint-Channels(TEJC)is proposed.On the one hand,it can solve the problem that the maximum stable extremum region(MSER)algorithm is only suitable for gray images and difficult to process color images.On the other hand,it solves the problem that the MSER algorithm has high complexity and low accuracy when extracting the most stable extreme region.In this paper,the proposed algorithm is tested and evaluated on the ICDAR data set.The experimental results show that the method has superiority.
基金This work was supported by the Sichuan Science and Technology Program(2021YFQ0003).
文摘Visual question answering(VQA)has attracted more and more attention in computer vision and natural language processing.Scholars are committed to studying how to better integrate image features and text features to achieve better results in VQA tasks.Analysis of all features may cause information redundancy and heavy computational burden.Attention mechanism is a wise way to solve this problem.However,using single attention mechanism may cause incomplete concern of features.This paper improves the attention mechanism method and proposes a hybrid attention mechanism that combines the spatial attention mechanism method and the channel attention mechanism method.In the case that the attention mechanism will cause the loss of the original features,a small portion of image features were added as compensation.For the attention mechanism of text features,a selfattention mechanism was introduced,and the internal structural features of sentences were strengthened to improve the overall model.The results show that attention mechanism and feature compensation add 6.1%accuracy to multimodal low-rank bilinear pooling network.
基金supported by the Science and Tec hnology Research and Development Plan Contract of China National Railway Group Co.,Ltd(Grant No.N2022G012)the Railway Science and Technology Research and Development Center Project(Project No.SYF2022SJ004).
文摘The robust guarantee of train control on-board equipment is inextricably linked to the safe functioning of a high-speed train.A fault diagnostic model of on-board equipment is built utilizing the integrated learning XGBoost(eXtreme Gradient Boosting)algorithm to help technicians assess the malfunction category of high-speed train control on-board equipment accurately and rapidly.The XGBoost algorithm iterates multiple decision tree models to improve the accuracy of fault diagnosis by lifting the predicted residual and adding regular terms.To begin,the text features were extracted using the improved TF-IDF(Term Frequency-Inverse Document Frequency)approach,and 24 fault feature words were chosen and converted into weight word vectors.Secondly,considering the imbalanced fault categories in the data set,the ADASYN(Adaptive Synthetic sampling)adaptive synthetically oversampling technique was used to synthesize a few category fault samples.Finally,the data samples were split into training and test sets based on the fault text data of CTCS-3train control on-board equipment recorded by Guangzhou Railway Group maintenance personnel.The XGBoost model was utilized to realize the automatic fault location of the test set after optimized parameter tuning through grid search.Compared with other methods,the evaluation index of the XGBoost model was significantly improved.The diagnostic accuracy reached 95.43%,which verifies the effectiveness of the method in text fault diagnosis.