Search

Text detection in complex scene images is a challenging task for intelligent transportation. Recently, anchor mechanisms are widely utilized in scene text detection tasks. However, in existing methods, anchors are generally predefined empirically, degrading robustness to complex scenarios with various sizes and orientation variations. In this paper, we propose a novel Attention Anchor Mechanism (AAM), especially targeting at predicting appropriate anchors for each pixel. To be concrete, we regard a series of predefined anchors as basic anchors and utilize an attention model to predict weights corresponding to basic anchors. Consequently, the weighted sum of basic anchors in each pixel can obtain a predicted anchor. In this way, the gap between the predicted anchors and the corresponding ground truth boxes could be narrowed, making the network easier to regress. For facilitating the design of basic anchors, we adopt a dimension-decomposition mechanism to predict width, height, and angle of anchors, respectively. Extensive experiments on several public datasets demonstrate that our method achieves state-of-the-art performance.

Existing multi-orientation scene text detection methods generally contain two crucial components: regression prediction for text bounding boxes and classification prediction for text/non-text. However, these methods always regard classification prediction and regression prediction as two independent procedures, neglecting fully exploring their mutual relations. Based on this key observation, we propose an innovative Scale-Guided Regression Module (SRM), specially for multi-orientation scene text detection. Equipped with width-guided kernels and height-guided kernels of different sizes, our SRM can generate a series of scale feature maps of candidate texts by capturing their shape information in classification prediction. The scale feature maps are used to predict the width and height of candidate texts, which can serve as guides for regressing bounding boxes. In this way, the procedures of classification and regression can be coherently integrated. In addition, we adopt IoU loss to train our network and then integrate IoU loss and l 1 -smooth loss for fine-tuning. Extensive experiments on publicly available datasets demonstrate the state-of-the-art performance of our method. Notably, our method achieves significant improvement of performance on long texts, e.g., on MSRA-TD500, our method outperforms Basemodel with a great margin (4.86 % in terms of Recall).

Anchor mechanism is widely applied in scene text detection methods and demonstrates promising performance. However, existing anchor mechanisms have two major limitations, namely handcrafted anchor design and hard-wired anchor assignment. We propose a novel Grouped Channels Composition(GCC) block to achieve the data-driven anchor design and adaptive anchor assignment. To be more specific, our GCC block uses optimizable anchor functions rather than handcrafted ones to achieve data-drive anchor design. In our GCC block, an adaptive anchor assignment is achieved with the attention mechanism instead of empirically assigning anchor according to the Intersection Over Union (IoU) between ground truth and targets. We then build a corresponding network named GCCNet with our GCC blocks. We also propose a Unified Loss Weighting module to alleviate the inconsistency between classification score and localization accuracy. Experiments conducted on publicly available datasets demonstrate the state-of-the-art performance of our methods.

Direct regression and anchor are the two mainly effective and prevailing mechanisms in the paradigm of scene text detection. However, the use of direct regression-based methods may be challenging during optimization without the help of anchors as references. Unfortunately, the anchor-based methods always suffer from the careful design of the anchors, degrading the robustness to complex scenes. To address the above-mentioned problems, we propose a novel hidden anchor mechanism (HAM) especially for scene text detection. The predictions of anchors are innovatively regarded as hidden layers, and the weighted sum of the predictions is integrated into a direct regression-based network. Hence, the architecture of our HAM still has the characteristic of simplicity as with direct regression-based methods. Moreover, it is easier to optimize anchors as references with this type of method than with direct regression-based methods. In this way, our network can take advantage of both direct regression and anchor mechanisms. In addition, we decouple three kinds of one-dimensional anchors from three-dimensional anchors, greatly reducing the number of anchors in text bounding box matching without performance degradation. We also propose a post-processing technique for long text detection, named iterative regression box (IRB), which takes a few additional computational costs and can be easily generalized to other methods. Experiments on several public datasets demonstrate that the proposed method achieves state-of-the-art performance. Code is available at https://github.com/hjbplayer/HAM .

Object detection is a significant and challenging problem in the study of remote sensing. Since remote sensing images are typically captured with a bird’s-eye view, the aspect ratios of objects in the same category may obey a Gaussian distribution. Generally, existing object detection methods ignore exploring the distribution character of aspect ratios for improving performance in remote sensing tasks. In this paper, we propose a novel Self-Adaptive Aspect Ratio Anchor (SARA) to explicitly explore aspect ratio variations of objects in remote sensing images. To be concrete, our SARA can self-adaptively learn an appropriate aspect ratio for each category. In this way, we can only utilize a simple squared anchor (related to the strides of feature maps in Feature Pyramid Networks) to regress objects in various aspect ratios. Finally, we adopt an Oriented Box Decoder (OBD) to align the feature maps and encode the orientation information of oriented objects. Our method achieves a promising mAP value of 79.91% on the DOTA dataset.

Arbitrary shape text detection is a challenging task due to the high variety and complexity of scenes texts. In this paper, we propose a novel unified relational reasoning graph network for arbitrary shape text detection. In our method, an innovative local graph bridges a text proposal model via Convolutional Neural Network (CNN) and a deep relational reasoning network via Graph Convolutional Network (GCN), making our network end-to-end trainable. To be concrete, every text instance will be divided into a series of small rectangular components, and the geometry attributes (e.g., height, width, and orientation) of the small components will be estimated by our text proposal model. Given the geometry attributes, the local graph construction model can roughly establish linkages between different text components. For further reasoning and deducing the likelihood of linkages between the component and its neighbors, we adopt a graph-based network to perform deep relational reasoning on local graphs. Experiments on public available datasets demonstrate the state-of-the-art performance of our method., Comment: 10 pages, Accepted by CVPR2020 Oral

Video has become one of the dominant data resources with the development of the Internet. As a result, the structured sorting of videos, which can be used for storage and extraction, represents a growing concern in the community. In particular, the text within videos can carry rich semantic information, leading to many novel studies wherein text tracking and recognition are performed. One essential step in text tracking involves template matching. In general, the adjacent matrices are modeled to represent the extracted tracking object features. Then, often, the Hungarian algorithm is applied to find the correspondence pairs between consecutive frames. In many works, text features are extracted based on morphological features such as color histograms and aspect ratios. However, under those features, similar text objects are not sufficiently distinguishable to make a distinction between them. To address this issue, we regard the template matching task as a graph matching problem. The main novelty involves a graph matching approach that utilizes the relationship between two trajectories or two objects, whereby a graph matching solver can be readily used in our tracking system. By utilizing the content information, the mismatch between the same object among different frames is effectively reduced. The experimental results demonstrate that the tracker with the graph matching method tends to increase the valid correspondence of trajectories and candidate objects.

Text extraction plays an important role in news images analysis tasks. Howerver, the conglutination of subtitles and station logos makes text detection challenging. In this paper, we develop an effective news text detection framework by introducing a novel similarity embedded proposal mechanism. The main idea is to predict similarity for each fine-scale coarse proposal to help construct text bounding boxes. Specifically, a CNN and bi-directional LSTM based network is used to produce vectors embedded in coarse proposals provided by Connectionist Text Proposal Network (CTPN). Notablely, similarity embedded proposal mechanism can be generalized to other sub-text level text detection models. Comparing to the state-of-the-art method (CTPN), our framework improves F-measure by 25.2% on our Private News Dataset and 8.9% on ICDAR 2013 benchmarks, respectively.

Based on the brief introduction of the variable frequency control technology, this paper elaborates the design and implementation of the variable frequency control strategy in the transmission line of the cementation facility, and describes the salient features of the control strategy. A frequency converter switches the control of two motors. And the reactor is used to achieve long-distance transmission of the frequency signal. In this project, the frequency regulation mainly affects the operation speed of the roller, and the shift of the roller speed is to take into account both the efficiency of waste treatment and the positioning accuracy of metal drum. The successful implementation of the project provides a good theoretical basis and example for the massive, standardized application of variable frequency control technology in the nuclear power plant.

Objective To establish a three–dimensional model of the temporal bone using CT scan images for study of temporal bone structures and simulation of mastoidectomy procedures. Methods CT scan images from 6 individuals (12 temporal bones) were used to reconstruct the Fallopian canal, internal auditory canal, cochlea, semicircular canals, sigmoid sinus, posterior fossa floor and jugular bulb on a computer platform. Their anatomical relations within the temporal bone were restored in the computed model. The same model was used to simulate mastoidectomy procedures. Results The reconstructed computer model provided accurate and clear three–dimensional images of temporal bone structures. Simulation of mastoidectomy using these images provided procedural experiences closely mimicking the real surgical procedure. Conclusion Computer–aided three dimensional reconstruction of temporal bone structures using CT scan images is a useful tool in surgical simulation and can aid surgical procedure planning. Key words three–dimension reconstruction, CT scan, surgery simulation