检索结果-内蒙古大学图书馆

Minimum description length clustering to measure meaningful image complexity

PATTERN RECOGNITION 2024年 145卷

作者： Mahon, Louis Lukasiewicz, Thomas Univ Edinburgh Sch Informat Edinburgh Scotland Vienna Univ Technol Inst Log & Computat Vienna Austria Univ Oxford Dept Comp Sci Oxford England

We present a new image complexity metric. Existing complexity metrics cannot distinguish meaningful content from noise, and give a high score to white noise images, which contain no meaningful information. We use the minimum description length principle to determine the number of clusters and designate certain points as outliers and, hence, correctly assign white noise a low score. The presented method is a step towards humans' ability to detect when data contain a meaningful pattern. It also has similarities to theoretical ideas for measuring meaningful complexity. We conduct experiments on seven different sets of images, which show that our method assigns the most accurate scores to all images considered. Additionally, comparing the different levels of the hierarchy of clusters can reveal how complexity manifests at different scales, from local detail to global structure. We then present ablation studies showing the contribution of the components of our method, and that it continues to assign reasonable scores when the inputs are modified in certain ways, including the addition of Gaussian noise and the lowering of the resolution. Code is available at https://***/Lou1sM/meaningful_image_complexity.

关键词： Meaningful complexity Clustering image complexity Minimum description length Machine learning Information theory

来源：评论

学校读者我要写书评

暂无评论

AENCIC: a method to estimate the number of clusters based on image complexity to be used in fuzzy clustering algorithms for image segmentation

引用

SOFT COMPUTING 2023年第15-16期28卷 8561页

作者： Madrid-Herrera, Luis Chacon-Murguia, Mario I. Ramirez-Quintana, Juan A. Tecnol Nacl Mexico IT Chihuahua Visual Percept Syst Lab Chihuahua Mexico

image segmentation through fuzzy clustering has been widely used in diverse areas. However, most of those clustering algorithms require that some of their parameter values be determined manually. The number of clusters, C, is one of the most important parameters because it impacts the number of regions to segment and directly affects the performance of the clustering algorithms. Some state-of-the-art general clustering algorithms methods automatically determine C. However, not all of them can be employed for image segmentation. Therefore, this paper describes the method automatic estimation of number of clusters by image complexity (AENCIC). AENCIC is a method that automatically estimates the best C needed by state-of-the-art clustering algorithms to segment an image, considering the image complexity perceived by humans. AENCIC was designed to work with fuzzy clustering algorithms employed to segment real-world images because this kind of segmentation is an ill-defined problem causing a high variation of C per image to attain a good segmentation. Results using the database BSDS500 demonstrate that using AENCIC to estimate C improves the performance of state-of-the-art fuzzy clustering image segmentation algorithms up to 94% of their ideal maximum performance, allowing those algorithms to work without human intervention.

关键词： Fuzzy image segmentation image complexity Fuzzy clustering Automatic clustering

来源：评论

学校读者我要写书评

暂无评论

A fuzzy approach to evaluate image segmentation based on image complexity

引用

IEEE LATIN AMERICA TRANSACTIONS 2023年第2期21卷 260-269页

作者： Madrid-Herrera, Luis Chacon-Murguia, Mario, I Ramirez-Quintana, Juan A. Tecnol Nacl Mexico IT Chihuahua Visual Percept Syst Lab Chihuahua 31310 Mexico

image segmentation evaluation may be carried out by comparing segmentation algorithm results with human ground truths. Its correct evaluation in benchmarks is important to promote the development of new and better segmentation algorithms. It may be partially considered a subjective task because each image may have multiple correct solutions. The evaluation is commonly carried out through crisp metrics, and they fail to generalize the subjectivity of the human criteria present in the human ground truths. Therefore, the interpretation and meaning of the metric values may be ambiguous. Thus, this paper presents a new fuzzy evaluation approach that considers the subjectivity of the human criteria by considering image complexity. This approach leads to an adequate evaluation of this subjective task. The proposed approach demonstrates its advantage when is evaluated using the Peng and SIHD meta-metrics achieving a performance of 0.795 and 0.971, respectively, outperforming the PRI, VI, GCE, and BDE metrics.

关键词： image segmentation Measurement complexity theory Task analysis Indexes IEEE transactions Visual perception fuzzy evaluation image complexity image segmentation evaluation

来源：评论

学校读者我要写书评

暂无评论

ICFRNet: image complexity Prior Guided Feature Refinement for Real-time Semantic Segmentation

ICFRNet: Image Complexity Prior Guided Feature Refinement fo...

引用

IEEE International Conference on Multimedia and Expo (ICME)

作者： Zhang, Xin Boyadzhiev, Teodor Shi, Jinglei Yang, Jufeng Nankai Univ Coll Comp Sci VCIP Tianjin Peoples R China Nankai Univ Coll Comp Sci TMCC Tianjin Peoples R China Nankai Univ Coll Comp Sci DISSec Tianjin Peoples R China Nankai Int Adv Res Inst SHENZHEN FUTIAN Shenzhen Peoples R China Bulgarian Acad Sci Inst Math & Informat Sofia Bulgaria

ISBN: (纸本)9798350390155;9798350390162

In this paper, we leverage image complexity as a prior for refining segmentation features to achieve accurate real-time semantic segmentation. The design philosophy is based on the observation that different pixel regions within an image exhibit varying levels of complexity, with higher complexities posing a greater challenge for accurate segmentation. We thus introduce image complexity as prior guidance and propose the image complexity prior-guided Feature Refinement Network (ICFRNet). This network aggregates both complexity and segmentation features to produce an attention map for refining segmentation features within an image complexity Guided Attention (ICGA) module. We optimize the network in terms of both segmentation and image complexity prediction tasks with a combined loss function. Experimental results on the Cityscapes and CamViD datasets have shown that our ICFRNet achieves higher accuracy with a competitive efficiency for real-time segmentation.

关键词： Real-time semantic segmentation image complexity dual-task framework

来源：评论

学校读者我要写书评

暂无评论

image complexity Guided Network Compression for Biomedical image Segmentation

引用

ACM JOURNAL ON EMERGING TECHNOLOGIES IN COMPUTING SYSTEMS 2022年第2期18卷 26-26页

作者： Mishra, Suraj Chen, Danny Z. Hu, X. Sharon Univ Notre Dame Dept Comp Sci & Engn Notre Dame IN 46556 USA

Compression is a standard procedure for making convolutional neural networks (CNNs) adhere to some specific computing resource constraints. However, searching for a compressed architecture typically involves a series of time-consuming training/validation experiments to determine a good compromise between network size and performance accuracy. To address this, we propose an image complexity-guided network compression technique for biomedical image segmentation. Given any resource constraints, our framework utilizes data complexity and network architecture to quickly estimate a compressed model which does not require network training. Specifically, we map the dataset complexity to the target network accuracy degradation caused by compression. Such mapping enables us to predict the final accuracy for different network sizes, based on the computed dataset complexity. Thus, one may choose a solution that meets both the network size and segmentation accuracy requirements. Finally, the mapping is used to determine the convolutional layer-wise multiplicative factor for generating a compressed network. We conduct experiments using 5 datasets, employing 3 commonly-used CNN architectures for biomedical image segmentation as representative networks. Our proposed framework is shown to be effective for generating compressed segmentation networks, retaining up to approximate to 95% of the full-sized network segmentation accuracy, and at the same time, utilizing approximate to 32x fewer network trainable weights (average reduction) of the full-sized networks.

关键词： Biomedical image segmentation convolutional neural networks network compression image complexity

来源：评论

学校读者我要写书评

暂无评论

image Visual complexity Evaluation Based on Deep Ordinal Regression 6th

Image Visual Complexity Evaluation Based on Deep Ordinal Reg...

引用

6th Chinese Conference on Pattern Recognition and Computer Vision (PRCV)

作者： Guo, Xiaoying Wang, Lu Yan, Tao Wei, Yanfeng Shanxi Univ Sch Automat & Software Engn Taiyuan Shanxi Peoples R China Shanxi Univ Inst Big Data Sci & Ind Taiyuan Shanxi Peoples R China Shanxi Univ Sch Compute & Informat Technol Taiyuan Shanxi Peoples R China Shanxi Univ Inst Management & Decis Sch Econ & Management Taiyuan Shanxi Peoples R China

ISBN: (纸本)9789819985517;9789819985524

image complexity is an important indicator in computer vision that helps people to more accurately evaluate and understand visual image information. Compared with traditional regression algorithms, ordinal regression methods are better suited to handling relationships and structures among ordinal data, providing more accurate reference for image complexity assessment. Currently, IC9600 dataset has made significant progress for providing the largest image complexity dataset, and ICNet provided a baseline model to evaluate the complexity score of images, but it neglects the ordinal property of complexity scores. This paper focuses on exploring a method to evaluate image complexity based on deep ordinal regression. We propose an evaluation model (ICCORN) that combines convolutional neural network ICNet and ordinal regression approach CORN. The model firstly extracts global semantic information and local detail information, and then considers ordinal relationship between complexity scores in the prediction process. The model demonstrates a high degree of correlation with human perception, as indicated by an increased Pearson correlation coefficient of 0.955. Furthermore, other evaluation metrics have also yielded favorable results.

关键词： image complexity complexity evaluation Ordinal regression Visual perception

来源：评论

学校读者我要写书评

暂无评论

Color image complexity versus Over-Segmentation: A Preliminary Study on the Correlation between complexity Measures and Number of Segments

引用

JOURNAL OF IMAGING 2020年第4期6卷 16-16页

作者： Ivanovici, Mihai Coliban, Radu-Mihai Hatfaludi, Cosmin Nicolae, Irina Emilia Transilvania Univ Brasov Elect & Comp Dept Brasov 500036 Romania

It is said that image segmentation is a very difficult or complex task. First of all, we emphasize the subtle difference between the notions of difficulty and complexity. Then, in this article, we focus on the question of how two widely used color image complexity measures correlate with the number of segments resulting in over-segmentation. We study the evolution of both the image complexity measures and number of segments as the image complexity is gradually decreased by means of low-pass filtering. In this way, we tackle the possibility of predicting the difficulty of color image segmentation based on image complexity measures. We analyze the complexity of images from the point of view of color entropy and color fractal dimension and for color fractal images and the Berkeley data set we correlate these two metrics with the segmentation results, more specifically the number of quasi-flat zones and the number of JSEG regions in the resulting segmentation map. We report on our experimental results and draw conclusions.

关键词： image complexity color entropy color fractal dimension quasi-flat zones JSEG

来源：评论

学校读者我要写书评

暂无评论

New approach of estimating edge detection threshold and application of adaptive detector depending on image complexity

引用

OPTIK 2021年 238卷 166476-166476页

作者： Maksimovic, Vladimir Petrovic, Mile Savic, Dragan Jaksic, Branimir Spalevic, Petar Univ Pristina Kosovska Mitrovica Fac Tech Sci Knjaza Milosa 7 Kosovska Mitrovica Serbia Singidunum Univ Fac Informat & Comp Danijelova 32 Belgrade Serbia

This paper presents a new approach to edge detection based on the application of an adequate adaptive algorithm for edge detection depending on the image complexity criteria. The complexity of the image was determined based on spatial information. Also, a new approach of estimating threshold for edge detection has been developed, which is based on the principles of machine learning, by applying grid search and random search. For analysis and testing, images with real scenes were used. This dataset was used to improve threshold estimation using the grid and random search methods. Based on the obtained results, it can be seen that the complexity significantly affects the edge detection, so with a low and medium number of details, the Roberts operator achieved good results. For images with a high number of details, the Canny operator has proven to be the best solution. New approach for estimate the threshold based on the grid search improves edge detection, especially with the Canny operator. Grid search-based method gives best results but it takes a lot of computational time. If computational time and the improvements obtained are taken into account, more effective optimization of the parameters for estimating threshold values is random search method.

关键词： Edge detection Grid search Random search image complexity Spatial information Threshold

来源：评论

学校读者我要写书评

暂无评论

Evaluation of image fire detection algorithms based on image complexity

引用

FIRE SAFETY JOURNAL 2021年 121卷 103306-103306页

作者： Li, Pu Yang, Yi Zhao, Wangda Zhang, Miao Zhengzhou Airport Econ Zone Fire & Rescue Div Zhengzhou 450000 Peoples R China Cent South Univ Sch Civil Engn Changsha 410075 Hunan Peoples R China Xian Univ Sci & Technol Xian 710054 Shaanxi Peoples R China Tianjin Heping Fire & Rescue Div Tianjin 300010 Peoples R China

Early fire detection is crucial for reducing losses due to fire. Therefore, the development of an efficient fire detection algorithm and alarm system is essential. The image fire detection algorithm is based on the mathematical analysis of images. Current algorithm evaluation methods are not effective and cannot sufficiently distinguish the performance of different detection algorithms. Therefore, such methods are not conducive for evaluating and improving algorithms. Based on a large-scale fire image dataset, the ground-truth complexity of images was quantified according to the time required for humans to detect the presence or absence of fire in the images. Four image complexity metrics based on the characteristics of fire detection are proposed. A comparison of the ground-truth and predicted scores of image complexity revealed that the comprehensive image complexity metric based on the Inception Resnet -v2 predictor was the most effective measurement. Its predicted scores ranked approximately 85% image pairs in the same order as that of the ground-truth complexity scores. Finally, a novel method for evaluating the performance of an image fire detection algorithm based on image complexity is proposed. Evaluation of the performance of five algorithms revealed that the performance of algorithm differs considerably and the proposed method can accurately determine the detection level of the detection algorithm in different image complexity conditions. The results of the study provide a valuable reference for developing and optimizing detection algorithms.

关键词： image complexity Fire detection Evaluate Algorithm Convolutional neural network

来源：评论

学校读者我要写书评

暂无评论

Evaluating Routine Variability of Daily Activities in Smart Homes with image complexity Measures

引用

JOURNAL OF COMPUTING IN CIVIL ENGINEERING 2020年第6期34卷

作者： Lee, Bogyeong Ahn, Changbum Ryan Mohan, Prakhar Chaspari, Theodora Lee, Hyun-Soo Texas A&M Univ Dept Construct Sci College Stn TX 77843 USA Texas A&M Univ Dept Comp Sci & Engn College Stn TX 77843 USA Seoul Natl Univ Dept Architecture & Architectural Engn Seoul 08826 South Korea

With the increasing trend of older adults living alone, an efficient and nonintrusive way to monitor these individuals' mental health status is required for early diagnosis of mental disease (e.g., dementia). Because the routine variability of activities of daily living (ADLs) can act as an index of an older adult's mental status, various research has attempted to develop a metric that can quantify and measure ADL routine variability. However, this research has focused either on the assessment of a single key ADL or the differences between activities performed on consecutive days. These approaches cannot measure the periodic changes over the long term (e.g., when the performed routine is different for each day of the week, or when exceptional events occurred) that may reflect mental health status. This study hypothesizes that the level of image complexity of the visualized data in ADL logs can represent the level of routine variability. To test this hypothesis, synthetic images are designed and generated presenting various randomness in ADL routines [i.e., varying the duration of routine-contributing activities (RADLs), varying the number of occurrences of non-routine-contributing activities (NRADLs), and varying the order of RADLs]. The correlations between each type of variability and the complexity values are identified, and values from canny edge distance and contrast of gray-level co-occurrence matrix (GLCM) elements show a strong correlation with the variability in the visualized data of the ADL logs. The metrics are further evaluated by using data collected from older adults living alone in real residential environments. This study suggests that image complexity metrics can be used to track gradual changes in routine variability within a subject, laying the foundation toward unobtrusive longitudinal measures of cognitive decline that could lead to the early prognosis of mental diseases.

关键词： Routine variability image complexity Activities of daily living Smart-home health care

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：