The main objective of this paper is to improve the image segmentation model for handwritten notebook analytics. We conducted a considerable amount of research in this area to increase the accuracy and efficiency of segmentation. To address the issues with traditional methods, we introduced attention mechanism and recursive residual convolutional neural network in the multi-task U-Net model. Through training and testing the model on handwritten notebook dataset and compared it with other existing technologies, we demonstrated the effectiveness of this method. The results showed that the model had a significant improvement in accuracy. Therefore, the research findings in this paper are important for improving the technology of handwritten notebook analytics.

Background subtraction networks outperform traditional hand-craft background subtraction methods. The main advantage of background subtraction networks is their ability to automatically learn background features for training scenes. When applying the trained network to new target scenes, adapting the network to the new scenes is crucial. However, few studies have focused on reusing multiple trained models for new target scenes. Considering background changes have several categories, such as illumination changes, a model trained for each background scene can work effectively for the target scene similar to the training scene. In this study, we propose a method to ensemble the module networks trained for each background scene. Experimental results show that the proposed method is significantly more accurate compared with the conventional methods in the target scene by tuning with only a few frames.

Handwritten notes are one important component of students’ learning process, which is used to record what they have learned in class or tease out knowledge after class for reflection and further strengthen the learning effect. It also helps a lot during review. We hope to divide handwritten notes (Japanese) into different parts, such as text, mathematical expressions, charts, etc., and quantify them to evaluate the condition of the notes and compare them among students. At the same time, data on students’ learning behaviors in the course are collected through the online education platform, such as the use time of textbook and attendance, as well as the scores of the online quiz and course grade. In this paper, the analysis of the relationship between the segmentation results of handwritten notes and learning behavior are reported, as well as the research on automatic page segmentation based on deep learning.

Abstract Automatic detection and counting of vehicles in a video is a challenging task and has become a key application area of traffic monitoring and management. In this paper, an efficient real-time approach for the detection and counting of moving vehicles is presented based on YOLOv2 and features point motion analysis. The work is based on synchronous vehicle features detection and tracking to achieve accurate counting results. The proposed strategy works in two phases; the first one is vehicle detection and the second is the counting of moving vehicles. Different convolutional neural networks including pixel by pixel classification networks and regression networks are investigated to improve the detection and counting decisions. For initial object detection, we have utilized state-of-the-art faster deep learning object detection algorithm YOLOv2 before refining them using K-means clustering and KLT tracker. Then an efficient approach is introduced using temporal information of the detection and tracking feature points between the framesets to assign each vehicle label with their corresponding trajectories and truly counted it. Experimental results on twelve challenging videos have shown that the proposed scheme generally outperforms state-of-the-art strategies. Moreover, the proposed approach using YOLOv2 increases the average time performance for the twelve tested sequences by 93.4% and 98.9% from 1.24 frames per second achieved using Faster Region-based Convolutional Neural Network (F R-CNN ) and 0.19 frames per second achieved using the background subtraction based CNN approach (BS-CNN ), respectively to 18.7 frames per second.

Recently, deep neural networks have demonstrated excellent performance in foreground segmentation tasks such as moving object detection and change detection tasks. Various types of neural networks have been proposed, however, the previous works mainly discuss the accuracy. Analytics of the neural networks is important to utilize them effectively and improve their performance. In this paper, we investigate a foreground segmentation network and background subtraction network. In our analysis, we discuss differences of behaviors of the two networks in specific scenes and feature distributions in each layer of a background subtraction network to investigate feature learning. In addition, we provide suggestions about the comparison with these networks.

Recently, background subtraction based on deep convolutional neural networks has demonstrated excellent performance in change detection tasks. However, most of the reported approaches require pixel-level label images for training the networks. To reduce the cost of rendering pixel-level annotation data, weakly supervised learning approaches using frame-level labels have been proposed. These labels indicate if a target class is present. Frame-level supervised learning is challenging because we cannot use location information for training the networks. Therefore, some constraints are introduced for guiding foreground locations. Previous works exploit prior information on foreground sizes and shapes. In this work, we propose two constraints for weakly supervised background subtraction networks. Our constraints use binary mask images generated by simple background subtraction. Unlike previous works, our approach does not require prior information on foreground sizes and shapes. Moreover, our constraints are more suitable for change detection tasks. We also present an experiment verifying that our constraints can improve foreground detection accuracy compared to other methods, which do not include them.

Recently, deep neural networks (DNNs) have demonstrated excellent performance for change detection. The DNN-based background subtraction automatically discovers background features from datasets and outperforms traditional background modeling based on handcraft features and/or subtraction strategies. Most researchers mainly discuss the accuracy of foreground detection and do not analyze how and why the DNN works well for change detection tasks. It is necessary to understand what the DNN learns as background features in order to discuss the potential of the DNN in background subtraction. In this paper, we focus on the filters in the first convolution layer and the activations of neurons in the last fully connected layer to understand the behavior of the DNN. From the experiment, we found that 1) the first layer performs the role of background subtraction using several filters, and 2) the last layer categorizes some background changes into a group without supervised signals. These findings suggest the possibility of a new background modeling strategy based on data-driven extracted features.

Background modeling and subtraction are essential to video surveillance applications. There are two main issues related to background modeling: how to initialize the background model, and how to update the model based on observations. In this paper, we consider the first issue with the aim of generating a clear background image that does not contain foreground objects or noise. We used a bidirectional analysis and consensus voting strategy to achieve this goal. We demonstrated the effectiveness of our technique using open access datasets.

We propose a strategy of background subtraction for an image sequence captured by a moving camera. To adapt for camera motion, it is necessary to estimate the relation between consecutive frames in background subtraction. However, simple background subtraction using the relation between consecutive frames results in many false detections. We use re-projection error to handle this problem. The re-projection error has a low value in a background region. According to re-projection error, our method searches neighboring background models and tunes a threshold value for detection in order to reduce false detections. We evaluated the accuracy of detection of our method in experiments. Our method provided better detection than a method that does not search neighboring background models. Our method thus reduced the number of false detections.