共查询到20条相似文献,搜索用时 15 毫秒
1.
This paper evaluates the degree of saliency of texts in natural scenes using visual saliency models. A large scale scene image database with pixel level ground truth is created for this purpose. Using this scene image database and five state-of-the-art models, visual saliency maps that represent the degree of saliency of the objects are calculated. The receiver operating characteristic curve is employed in order to evaluate the saliency of scene texts, which is calculated by visual saliency models. A visualization of the distribution of scene texts and non-texts in the space constructed by three kinds of saliency maps, which are calculated using Itti''s visual saliency model with intensity, color and orientation features, is given. This visualization of distribution indicates that text characters are more salient than their non-text neighbors, and can be captured from the background. Therefore, scene texts can be extracted from the scene images. With this in mind, a new visual saliency architecture, named hierarchical visual saliency model, is proposed. Hierarchical visual saliency model is based on Itti''s model and consists of two stages. In the first stage, Itti''s model is used to calculate the saliency map, and Otsu''s global thresholding algorithm is applied to extract the salient region that we are interested in. In the second stage, Itti''s model is applied to the salient region to calculate the final saliency map. An experimental evaluation demonstrates that the proposed model outperforms Itti''s model in terms of captured scene texts. 相似文献
2.
Saliency detection is widely used in many visual applications like image segmentation, object recognition and classification. In this paper, we will introduce a new method to detect salient objects in natural images. The approach is based on a regional principal color contrast modal, which incorporates low-level and medium-level visual cues. The method allows a simple computation of color features and two categories of spatial relationships to a saliency map, achieving higher F-measure rates. At the same time, we present an interpolation approach to evaluate resulting curves, and analyze parameters selection. Our method enables the effective computation of arbitrary resolution images. Experimental results on a saliency database show that our approach produces high quality saliency maps and performs favorably against ten saliency detection algorithms. 相似文献
3.
The automatic computerized detection of regions of interest (ROI) is an important step in the process of medical image processing and analysis. The reasons are many, and include an increasing amount of available medical imaging data, existence of inter-observer and inter-scanner variability, and to improve the accuracy in automatic detection in order to assist doctors in diagnosing faster and on time. A novel algorithm, based on visual saliency, is developed here for the identification of tumor regions from MR images of the brain. The GBM saliency detection model is designed by taking cue from the concept of visual saliency in natural scenes. A visually salient region is typically rare in an image, and contains highly discriminating information, with attention getting immediately focused upon it. Although color is typically considered as the most important feature in a bottom-up saliency detection model, we circumvent this issue in the inherently gray scale MR framework. We develop a novel pseudo-coloring scheme, based on the three MRI sequences, viz. FLAIR, T2 and T1C (contrast enhanced with Gadolinium). A bottom-up strategy, based on a new pseudo-color distance and spatial distance between image patches, is defined for highlighting the salient regions in the image. This multi-channel representation of the image and saliency detection model help in automatically and quickly isolating the tumor region, for subsequent delineation, as is necessary in medical diagnosis. The effectiveness of the proposed model is evaluated on MRI of 80 subjects from the BRATS database in terms of the saliency map values. Using ground truth of the tumor regions for both high- and low- grade gliomas, the results are compared with four highly referred saliency detection models from literature. In all cases the AUC scores from the ROC analysis are found to be more than 0.999 ± 0.001 over different tumor grades, sizes and positions. 相似文献
4.
The detection of airport attracts lots of attention and becomes a hot topic recently because of its applications and importance in military and civil aviation fields. However, the complicated background around airports brings much difficulty into the detection. This paper presents a new method for airport detection in remote sensing images. Distinct from other methods which analyze images pixel by pixel, we introduce visual attention mechanism into detection of airport and improve the efficiency of detection greatly. Firstly, Hough transform is used to judge whether an airport exists in an image. Then an improved graph-based visual saliency model is applied to compute the saliency map and extract regions of interest (ROIs). The airport target is finally detected according to the scale-invariant feature transform features which are extracted from each ROI and classified by hierarchical discriminant regression tree. Experimental results show that the proposed method is faster and more accurate than existing methods, and has lower false alarm rate and better anti-noise performance simultaneously. 相似文献
5.
Humans have been shown to combine noisy sensory information with previous experience (priors), in qualitative and sometimes quantitative agreement with the statistically-optimal predictions of Bayesian integration. However, when the prior distribution becomes more complex than a simple Gaussian, such as skewed or bimodal, training takes much longer and performance appears suboptimal. It is unclear whether such suboptimality arises from an imprecise internal representation of the complex prior, or from additional constraints in performing probabilistic computations on complex distributions, even when accurately represented. Here we probe the sources of suboptimality in probabilistic inference using a novel estimation task in which subjects are exposed to an explicitly provided distribution, thereby removing the need to remember the prior. Subjects had to estimate the location of a target given a noisy cue and a visual representation of the prior probability density over locations, which changed on each trial. Different classes of priors were examined (Gaussian, unimodal, bimodal). Subjects'' performance was in qualitative agreement with the predictions of Bayesian Decision Theory although generally suboptimal. The degree of suboptimality was modulated by statistical features of the priors but was largely independent of the class of the prior and level of noise in the cue, suggesting that suboptimality in dealing with complex statistical features, such as bimodality, may be due to a problem of acquiring the priors rather than computing with them. We performed a factorial model comparison across a large set of Bayesian observer models to identify additional sources of noise and suboptimality. Our analysis rejects several models of stochastic behavior, including probability matching and sample-averaging strategies. Instead we show that subjects'' response variability was mainly driven by a combination of a noisy estimation of the parameters of the priors, and by variability in the decision process, which we represent as a noisy or stochastic posterior. 相似文献
6.
Psychophysical tests of the hypothesis of a bottom-up saliency map in primary visual cortex
下载免费PDF全文
![点击此处可从《PLoS computational biology》网站下载免费的PDF全文](/ch/ext_images/free.gif)
A unique vertical bar among horizontal bars is salient and pops out perceptually. Physiological data have suggested that mechanisms in the primary visual cortex (V1) contribute to the high saliency of such a unique basic feature, but indicated little regarding whether V1 plays an essential or peripheral role in input-driven or bottom-up saliency. Meanwhile, a biologically based V1 model has suggested that V1 mechanisms can also explain bottom-up saliencies beyond the pop-out of basic features, such as the low saliency of a unique conjunction feature such as a red vertical bar among red horizontal and green vertical bars, under the hypothesis that the bottom-up saliency at any location is signaled by the activity of the most active cell responding to it regardless of the cell's preferred features such as color and orientation. The model can account for phenomena such as the difficulties in conjunction feature search, asymmetries in visual search, and how background irregularities affect ease of search. In this paper, we report nontrivial predictions from the V1 saliency hypothesis, and their psychophysical tests and confirmations. The prediction that most clearly distinguishes the V1 saliency hypothesis from other models is that task-irrelevant features could interfere in visual search or segmentation tasks which rely significantly on bottom-up saliency. For instance, irrelevant colors can interfere in an orientation-based task, and the presence of horizontal and vertical bars can impair performance in a task based on oblique bars. Furthermore, properties of the intracortical interactions and neural selectivities in V1 predict specific emergent phenomena associated with visual grouping. Our findings support the idea that a bottom-up saliency map can be at a lower visual area than traditionally expected, with implications for top-down selection mechanisms. 相似文献
7.
Coral reefs are rich in fisheries and aquatic resources, and the study and monitoring of coral reef ecosystems are of great economic value and practical significance. Due to complex backgrounds and low-quality videos, it is challenging to identify coral reef fish. This study proposed an image enhancement approach for fish detection in complex underwater environments. The method first uses a Siamese network to obtain a saliency map and then multiplies this saliency map by the input image to construct an image enhancement module. Applying this module to the existing mainstream one-stage and two-stage target detection frameworks can significantly improve their detection accuracy. Good detection performance was achieved in a variety of scenarios, such as those with luminosity variations, aquatic plant movements, blurred images, large targets and multiple targets, demonstrating the robustness of the algorithm. The best performance was achieved on the LCF-15 dataset when combining the proposed method with the cascade region-based convolutional neural network (Cascade-RCNN). The average precision at an intersection-over-union (IoU) threshold of 0.5 (AP50) was 0.843, and the F1 score was 0.817, exceeding the best reported results on this dataset. This study provides an automated video analysis tool for marine-related researchers and technical support for downstream applications. 相似文献
8.
Hill SM Neve RM Bayani N Kuo WL Ziyad S Spellman PT Gray JW Mukherjee S 《BMC bioinformatics》2012,13(1):94
ABSTRACT: BACKGROUND: An important question in the analysis of biochemical data is that of identifying subsets of molecular variables that may jointly influence a biological response. Statistical variable selection methods have been widely used for this purpose. In many settings, it may be important to incorporate ancillary biological information concerning the variables of interest. Pathway and network maps are one example of a source of such information. However, although ancillary information is increasingly available, it is not always clear how it should be used nor how it should be weighted in relation to primary data. RESULTS: We put forward an approach in which biological knowledge is incorporated using informative prior distributions over variable subsets, with prior information selected and weighted in an automated, objective manner using an empirical Bayes formulation. We employ continuous, linear models with interaction terms and exploit biochemically-motivated sparsity constraints to permit exact inference. We show an example of priors for pathway- and network-based information and illustrate our proposed method on both synthetic response data and by an application to cancer drug response data. Comparisons are also made to alternative Bayesian and frequentist penalised-likelihood methods for incorporating network-based information. CONCLUSIONS: The empirical Bayes method proposed here can aid prior elicitation for Bayesian variable selection studies and help to guard against mis-specification of priors. Empirical Bayes, together with the proposed pathway-based priors, results in an approach with a competitive variable selection performance. In addition, the overall procedure is fast, deterministic, and has very few user-set parameters, yet is capable of capturing interplay between molecular players. The approach presented is general and readily applicable in any setting with multiple sources of biological prior knowledge. 相似文献
9.
The neglected tool in the Bayesian ecologist's shed: a case study testing informative priors' effect on model accuracy
下载免费PDF全文
![点击此处可从《Ecology and evolution》网站下载免费的PDF全文](/ch/ext_images/free.gif)
William K. Morris Peter A. Vesk Michael A. McCarthy Sarayudh Bunyavejchewin Patrick J. Baker 《Ecology and evolution》2015,5(1):102-108
Despite benefits for precision, ecologists rarely use informative priors. One reason that ecologists may prefer vague priors is the perception that informative priors reduce accuracy. To date, no ecological study has empirically evaluated data‐derived informative priors' effects on precision and accuracy. To determine the impacts of priors, we evaluated mortality models for tree species using data from a forest dynamics plot in Thailand. Half the models used vague priors, and the remaining half had informative priors. We found precision was greater when using informative priors, but effects on accuracy were more variable. In some cases, prior information improved accuracy, while in others, it was reduced. On average, models with informative priors were no more or less accurate than models without. Our analyses provide a detailed case study on the simultaneous effect of prior information on precision and accuracy and demonstrate that when priors are specified appropriately, they lead to greater precision without systematically reducing model accuracy. 相似文献
10.
Rufin VanRullen 《Journal of Physiology》2003,97(2-3):365-377
Visual saliency is a fundamental yet hard to define property of objects or locations in the visual world. In a context where objects and their representations compete to dominate our perception, saliency can be thought of as the "juice" that makes objects win the race. It is often assumed that saliency is extracted and represented in an explicit saliency map, which serves to determine the location of spatial attention at any given time. It is then by drawing attention to a salient object that it can be recognized or categorized. I argue against this classical view that visual "bottom-up" saliency automatically recruits the attentional system prior to object recognition. A number of visual processing tasks are clearly performed too fast for such a costly strategy to be employed. Rather, visual attention could simply act by biasing a saliency-based object recognition system. Under natural conditions of stimulation, saliency can be represented implicitly throughout the ventral visual pathway, independent of any explicit saliency map. At any given level, the most activated cells of the neural population simply represent the most salient locations. The notion of saliency itself grows increasingly complex throughout the system, mostly based on luminance contrast until information reaches visual cortex, gradually incorporating information about features such as orientation or color in primary visual cortex and early extrastriate areas, and finally the identity and behavioral relevance of objects in temporal cortex and beyond. Under these conditions the object that dominates perception, i.e. the object yielding the strongest (or the first) selective neural response, is by definition the one whose features are most "salient"--without the need for any external saliency map. In addition, I suggest that such an implicit representation of saliency can be best encoded in the relative times of the first spikes fired in a given neuronal population. In accordance with our subjective experience that saliency and attention do not modify the appearance of objects, the feed-forward propagation of this first spike wave could serve to trigger saliency-based object recognition outside the realm of awareness, while conscious perceptions could be mediated by the remaining discharges of longer neuronal spike trains. 相似文献
11.
Many saliency computational models have been proposed to simulate bottom-up visual attention mechanism of human visual system. However, most of them only deal with certain kinds of images or aim at specific applications. In fact, human beings have the ability to correctly select attentive focuses of objects with arbitrary sizes within any scenes. This paper proposes a new bottom-up computational model from the perspective of frequency domain based on the biological discovery of non-Classical Receptive Field (nCRF) in the retina. A saliency map can be obtained according to the idea of Extended Classical Receptive Field. The model is composed of three major steps: firstly decompose the input image into several feature maps representing different frequency bands that cover the whole frequency domain by utilizing Gabor wavelet. Secondly, whiten the feature maps to highlight the embedded saliency information. Thirdly, select some optimal maps, simulating the response of receptive field especially nCRF, to generate the saliency map. Experimental results show that the proposed algorithm is able to work with stable effect and outstanding performance in a variety of situations as human beings do and is adaptive to both psychological patterns and natural images. Beyond that, biological plausibility of nCRF and Gabor wavelet transform make this approach reliable. 相似文献
12.
In this paper we propose a computational model of bottom–up visual attention based on a pulsed principal component analysis
(PCA) transform, which simply exploits the signs of the PCA coefficients to generate spatial and motional saliency. We further
extend the pulsed PCA transform to a pulsed cosine transform that is not only data-independent but also very fast in computation.
The proposed model has the following biological plausibilities. First, the PCA projection vectors in the model can be obtained
by using the Hebbian rule in neural networks. Second, the outputs of the pulsed PCA transform, which are inherently binary,
simulate the neuronal pulses in the human brain. Third, like many Fourier transform-based approaches, our model also accomplishes
the cortical center-surround suppression in frequency domain. Experimental results on psychophysical patterns and natural
images show that the proposed model is more effective in saliency detection and predict human eye fixations better than the
state-of-the-art attention models. 相似文献
13.
In recent years, there has been considerable interest in visual attention models (saliency map of visual attention). These models can be used to predict eye fixation locations, and thus will have many applications in various fields which leads to obtain better performance in machine vision systems. Most of these models need to be improved because they are based on bottom-up computation that does not consider top-down image semantic contents and often does not match actual eye fixation locations. In this study, we recorded the eye movements (i.e., fixations) of fourteen individuals who viewed images which consist natural (e.g., landscape, animal) and man-made (e.g., building, vehicles) scenes. We extracted the fixation locations of eye movements in two image categories. After extraction of the fixation areas (a patch around each fixation location), characteristics of these areas were evaluated as compared to non-fixation areas. The extracted features in each patch included the orientation and spatial frequency. After feature extraction phase, different statistical classifiers were trained for prediction of eye fixation locations by these features. This study connects eye-tracking results to automatic prediction of saliency regions of the images. The results showed that it is possible to predict the eye fixation locations by using of the image patches around subjects’ fixation points. 相似文献
14.
Meysam Hashemi Anirudh N. Vattikonda Viktor Sip Sandra Diaz-Pier Alexander Peyser Huifang Wang Maxime Guye Fabrice Bartolomei Marmaduke M. Woodman Viktor K. Jirsa 《PLoS computational biology》2021,17(7)
Individualized anatomical information has been used as prior knowledge in Bayesian inference paradigms of whole-brain network models. However, the actual sensitivity to such personalized information in priors is still unknown. In this study, we introduce the use of fully Bayesian information criteria and leave-one-out cross-validation technique on the subject-specific information to assess different epileptogenicity hypotheses regarding the location of pathological brain areas based on a priori knowledge from dynamical system properties. The Bayesian Virtual Epileptic Patient (BVEP) model, which relies on the fusion of structural data of individuals, a generative model of epileptiform discharges, and a self-tuning Monte Carlo sampling algorithm, is used to infer the spatial map of epileptogenicity across different brain areas. Our results indicate that measuring the out-of-sample prediction accuracy of the BVEP model with informative priors enables reliable and efficient evaluation of potential hypotheses regarding the degree of epileptogenicity across different brain regions. In contrast, while using uninformative priors, the information criteria are unable to provide strong evidence about the epileptogenicity of brain areas. We also show that the fully Bayesian criteria correctly assess different hypotheses about both structural and functional components of whole-brain models that differ across individuals. The fully Bayesian information-theory based approach used in this study suggests a patient-specific strategy for epileptogenicity hypothesis testing in generative brain network models of epilepsy to improve surgical outcomes. 相似文献
15.
Multispecies occupancy models can estimate species richness from spatially replicated multispecies detection/non‐detection survey data, while accounting for imperfect detection. A model extension using data augmentation allows inferring the total number of species in the community, including those completely missed by sampling (i.e., not detected in any survey, at any site). Here we investigate the robustness of these estimates. We review key model assumptions and test performance via simulations, under a range of scenarios of species characteristics and sampling regimes, exploring sensitivity to the Bayesian priors used for model fitting. We run tests when assumptions are perfectly met and when violated. We apply the model to a real dataset and contrast estimates obtained with and without predictors, and for different subsets of data. We find that, even with model assumptions perfectly met, estimation of the total number of species can be poor in scenarios where many species are missed (>15%–20%) and that commonly used priors can accentuate overestimation. Our tests show that estimation can often be robust to violations of assumptions about the statistical distributions describing variation of occupancy and detectability among species, but lower‐tail deviations can result in large biases. We obtain substantially different estimates from alternative analyses of our real dataset, with results suggesting that missing relevant predictors in the model can result in richness underestimation. In summary, estimates of total richness are sensitive to model structure and often uncertain. Appropriate selection of priors, testing of assumptions, and model refinement are all important to enhance estimator performance. Yet, these do not guarantee accurate estimation, particularly when many species remain undetected. While statistical models can provide useful insights, expectations about accuracy in this challenging prediction task should be realistic. Where knowledge about species numbers is considered truly critical for management or policy, survey effort should ideally be such that the chances of missing species altogether are low. 相似文献
16.
17.
Xing EP Wu W Jordan MI Karp RM 《Journal of bioinformatics and computational biology》2004,2(1):127-154
The complexity of the global organization and internal structure of motifs in higher eukaryotic organisms raises significant challenges for motif detection techniques. To achieve successful de novo motif detection, it is necessary to model the complex dependencies within and among motifs and to incorporate biological prior knowledge. In this paper, we present LOGOS, an integrated LOcal and GlObal motif Sequence model for biopolymer sequences, which provides a principled framework for developing, modularizing, extending and computing expressive motif models for complex biopolymer sequence analysis. LOGOS consists of two interacting submodels: HMDM, a local alignment model capturing biological prior knowledge and positional dependency within the motif local structure; and HMM, a global motif distribution model modeling frequencies and dependencies of motif occurrences. Model parameters can be fit using training motifs within an empirical Bayesian framework. A variational EM algorithm is developed for de novo motif detection. LOGOS improves over existing models that ignore biological priors and dependencies in motif structures and motif occurrences, and demonstrates superior performance on both semi-realistic test data and cis-regulatory sequences from yeast and Drosophila genomes with regard to sensitivity, specificity, flexibility and extensibility. 相似文献
18.
Wrong,but useful: regional species distribution models may not be improved by range‐wide data under biased sampling
下载免费PDF全文
![点击此处可从《Ecology and evolution》网站下载免费的PDF全文](/ch/ext_images/free.gif)
Species distribution modeling (SDM) is an essential method in ecology and conservation. SDMs are often calibrated within one country's borders, typically along a limited environmental gradient with biased and incomplete data, making the quality of these models questionable. In this study, we evaluated how adequate are national presence‐only data for calibrating regional SDMs. We trained SDMs for Egyptian bat species at two different scales: only within Egypt and at a species‐specific global extent. We used two modeling algorithms: Maxent and elastic net, both under the point‐process modeling framework. For each modeling algorithm, we measured the congruence of the predictions of global and regional models for Egypt, assuming that the lower the congruence, the lower the appropriateness of the Egyptian dataset to describe the species' niche. We inspected the effect of incorporating predictions from global models as additional predictor (“prior”) to regional models, and quantified the improvement in terms of AUC and the congruence between regional models run with and without priors. Moreover, we analyzed predictive performance improvements after correction for sampling bias at both scales. On average, predictions from global and regional models in Egypt only weakly concur. Collectively, the use of priors did not lead to much improvement: similar AUC and high congruence between regional models calibrated with and without priors. Correction for sampling bias led to higher model performance, whatever prior used, making the use of priors less pronounced. Under biased and incomplete sampling, the use of global bats data did not improve regional model performance. Without enough bias‐free regional data, we cannot objectively identify the actual improvement of regional models after incorporating information from the global niche. However, we still believe in great potential for global model predictions to guide future surveys and improve regional sampling in data‐poor regions. 相似文献
19.
Jamie R. Oaks Charles W. Linkem Jeet Sukumaran 《Evolution; international journal of organic evolution》2014,68(12):3607-3617
Establishing that a set of population‐splitting events occurred at the same time can be a potentially persuasive argument that a common process affected the populations. Recently, Oaks et al. ( 2013 ) assessed the ability of an approximate‐Bayesian model‐choice method (msBayes ) to estimate such a pattern of simultaneous divergence across taxa, to which Hickerson et al. ( 2014 ) responded. Both papers agree that the primary inference enabled by the method is very sensitive to prior assumptions and often erroneously supports shared divergences across taxa when prior uncertainty about divergence times is represented by a uniform distribution. However, the papers differ about the best explanation and solution for this problem. Oaks et al. ( 2013 ) suggested the method's behavior was caused by the strong weight of uniformly distributed priors on divergence times leading to smaller marginal likelihoods (and thus smaller posterior probabilities) of models with more divergence‐time parameters (Hypothesis 1); they proposed alternative prior probability distributions to avoid such strongly weighted posteriors. Hickerson et al. ( 2014 ) suggested numerical‐approximation error causes msBayes analyses to be biased toward models of clustered divergences because the method's rejection algorithm is unable to adequately sample the parameter space of richer models within reasonable computational limits when using broad uniform priors on divergence times (Hypothesis 2). As a potential solution, they proposed a model‐averaging approach that uses narrow, empirically informed uniform priors. Here, we use analyses of simulated and empirical data to demonstrate that the approach of Hickerson et al. ( 2014 ) does not mitigate the method's tendency to erroneously support models of highly clustered divergences, and is dangerous in the sense that the empirically derived uniform priors often exclude from consideration the true values of the divergence‐time parameters. Our results also show that the tendency of msBayes analyses to support models of shared divergences is primarily due to Hypothesis 1, whereas Hypothesis 2 is an untenable explanation for the bias. Overall, this series of papers demonstrates that if our prior assumptions place too much weight in unlikely regions of parameter space such that the exact posterior supports the wrong model of evolutionary history, no amount of computation can rescue our inference. Fortunately, as predicted by fundamental principles of Bayesian model choice, more flexible distributions that accommodate prior uncertainty about parameters without placing excessive weight in vast regions of parameter space with low likelihood increase the method's robustness and power to detect temporal variation in divergences. 相似文献
20.
The human visual system uses texture information to automatically, or pre-attentively, segregate parts of the visual scene. We investigate the neural substrate underlying human texture processing using a computational model that consists of a hierarchy of bi-directionally linked model areas. The model builds upon two key hypotheses, namely that (i) texture segregation is based on boundary detection--rather than clustering of homogeneous items--and (ii) texture boundaries are detected mainly on the basis of a large scenic context that is analyzed by higher cortical areas within the ventral visual pathway, such as area V4. Here, we focus on the interpretation of key results from psychophysical studies on human texture segmentation. In psychophysical studies, texture patterns were varied along several feature dimensions to systematically characterize human performance. We use simulations to demonstrate that the activation patterns of our model directly correlate with the psychophysical results. This allows us to identify the putative neural mechanisms and cortical key areas which underlie human behavior. In particular, we investigate (i) the effects of varying texture density on target saliency, and the impact of (ii) element alignment and (iii) orientation noise on the detectability of a pop-out bar. As a result, we demonstrate that the dependency of target saliency on texture density is linked to a putative receptive field organization of orientation-selective neurons in V4. The effect of texture element alignment is related to grouping mechanisms in early visual areas. Finally, the modulation of cell activity by feedback activation from higher model areas, interacting with mechanisms of intra-areal center-surround competition, is shown to result in the specific suppression of noise-related cell activities and to improve the overall model capabilities in texture segmentation. In particular, feedback interaction is crucial to raise the model performance to the level of human observers. 相似文献