Academic literature on the topic 'Perceptual dissimilarity meaure'

It has been shown that visual display systems of intonation can be employed beneficially in teaching intonation to persons with deafness and in teaching the intonation of a foreign language. In current training situations the correctness of a reproduced pitch contour is rated either by the teacher or automatically. In the latter case an algorithm mostly estimates the maximum deviation from an example contour. In game-like exercises, for instance, the pupil has to produce a pitch contour within the displayed floor and ceiling of a "tunnel" with a preadjusted height. In an experiment described in the companion paper, phoneticians had rated the dissimilarity of two pitch contours both auditorily, by listening to two resynthesized utterances, and visually, by looking at two pitch contours displayed on a computer screen. A test is reported in which these dissimilarity ratings were compared with automatic ratings obtained with this tunnel measure and with three other measures, the mean distance, the root-mean-square (RMS) distance, and the correlation coefficient. The most frequently used tunnel measure appeared to have the weakest correlation with the ratings by the phoneticians. In general, the automatic ratings obtained with the correlation coefficient showed the strongest correlation with the perceptual ratings. A disadvantage of this measure, however, may be that it normalizes for the range of the pitch contours. If range is important, as in intonation teaching to persons with deafness, the mean distance or the RMS distance are the best physical measures for automatic training of intonation.

1. Using operant techniques, rats were trained to differentially report (i.e., identify) the odorants propanol, carvone, citral, propyl acetate, and ethylacetoacetate. After acquisition training, the animals were tested using a 5 x 5 confusion matrix design. The results of the behavioral tests were used to measure the degree of perceptual dissimilarity between any pair of odorants. These dissimilarity measures were then subjected to multidimensional scaling analysis to establish a two-dimensional perceptual odor space for each rat. 2. At the completion of behavioral testing, the fluorescence changes in the dye di-4-ANEPPS were monitored on the rat's nasal septum and medial surface of the turbinates in response to the same odorants. For each mucosal surface a 6.0 x 6.0 mm area was sampled at 100 contiguous sites with a 10 x 10 photodiode array. 3. Formal statistical analysis indicated a highly significant predictive relationship between the relative position of an odorant's mucosal loci of maximal activity or “hot spot” and the relative position of the same odorant in a psychophysically determined perceptual odor space (F = 15.6, P < 0.001). 4. The results of this study suggest for the first time that odorant-induced mucosal activity patterns serve as the substrate for the perception of odorant quality.

Although crosslinguistic similarity is a crucial concept for many disciplines in the speech sciences, there is no clear consensus as to the most appropriate method to measure it. This paper assessed the perceived similarity between English and Catalan vowels by means of an overt direct task evaluating perceptual similarity. The extent to which perceptual similarity is reciprocal is also explored by comparing perceptual judgements obtained by speakers of the two languages involved. Twenty-seven native Catalan speakers and 27 native English speakers rated the perceived dissimilarity between two aurally presented vowel stimuli. Trials included native–non-native pairs as well as native-native pairs to serve as baseline data. Some native–non-native pairs were perceived to be as similar as same-category native pairs, illustrating cases of very high crosslinguistic perceptual similarity. Further, in terms of reciprocity, the results showed a bidirectionality in similarity relationships that point to some cases of near-identical or shared categories and also illustrate the role of language-specific cue weighting in determining perceptual similarity. Finally, a comparison with the outcome of a previous study [Cebrian (2021). J. Acoust. Soc. Am. 149(4), 2671–2685], involving the same participants and languages but exploring ecphoric similarity, shows a generally high degree of agreement and a close relationship between the two types of similarity.

We propose a texture-based interpretation workflow and apply it to delineate salt domes in 3D migrated seismic volumes. First, we compute an attribute map using a novel seismic attribute, 3D gradient of textures (3D-GoT), which measures the dissimilarity between neighboring cubes around each voxel in a seismic volume across the time or depth, crossline, and inline directions. To evaluate the texture dissimilarity, we introduce five 3D perceptual and nonperceptual dissimilarity functions. Second, we apply a global threshold on the 3D-GoT volume to yield a binary volume and demonstrate its effects on salt-dome delineation using objective evaluation measures such as receiver operating characteristic curves and the areas under the curves. Third, with an initial seed point selected inside the binary volume, we use a 3D region growing method to capture a salt body. For an automated 3D region growing, we adopt a tensor-based automatic seed point selection method. Finally, we apply morphological postprocessing to delineate the salt dome within the seismic volume. Furthermore, we also develop an objective evaluation measure based on the curvature and shape to compute the similarity between detected salt-dome boundaries and the reference interpreted by the geophysicist. Experimental results on a real data set from the North Sea show that the proposed method outperforms the state-of-the-art methods for salt-dome delineation.

Generalized flash suppression (GFS), in which a salient visual stimulus can be rendered invisible despite continuous retinal input, provides a rare opportunity to directly study the neural mechanism of visual perception. Previous work based on linear methods, such as spectral analysis, on local field potential (LFP) during GFS has shown that the LFP power at distinctive frequency bands are differentially modulated by perceptual suppression. Yet, the linear method alone may be insufficient for the full assessment of neural dynamic due to the fundamentally nonlinear nature of neural signals. In this study, we set forth to analyze the LFP data collected from multiple visual areas in V1, V2 and V4 of macaque monkeys while performing the GFS task using a nonlinear method — adaptive multi-scale entropy (AME) — to reveal the neural dynamic of perceptual suppression. In addition, we propose a new cross-entropy measure at multiple scales, namely adaptive multi-scale cross-entropy (AMCE), to assess the nonlinear functional connectivity between two cortical areas. We show that: (1) multi-scale entropy exhibits percept-related changes in all three areas, with higher entropy observed during perceptual suppression; (2) the magnitude of the perception-related entropy changes increases systematically over successive hierarchical stages (i.e. from lower areas V1 to V2, up to higher area V4); and (3) cross-entropy between any two cortical areas reveals higher degree of asynchrony or dissimilarity during perceptual suppression, indicating a decreased functional connectivity between cortical areas. These results, taken together, suggest that perceptual suppression is related to a reduced functional connectivity and increased uncertainty of neural responses, and the modulation of perceptual suppression is more effective at higher visual cortical areas. AME is demonstrated to be a useful technique in revealing the underlying dynamic of nonlinear/nonstationary neural signal.

Earth mover's distance (EMD for short) is a perceptually meaningful dissimilarity measure between histograms. The computation of EMD reduces to a network flow optimization problem; however, it lays a heavy computational burden when the number of locations of histograms is large. In this paper, we address an efficient formulation for computing the exact EMD value. We prove that the EMD problem reduces to a problem with half the number of constraints regardless of the ground distance. We then propose a further reduced formula in which the number of variables is reduced from O(m2) to O(m) for histograms with m locations when the ground distance is derived from a graph with a homogeneous neighborhood structure. Specifically, EMD problems with L1, L∞ and D-norm ground distances can be reduced in this manner. Some experiments show that the reduction helps compute the EMD efficiently.

Image retrieval and clustering are two important tools for analysing and organising images. Dissimilarity measure is central to both image retrieval and clustering. The performance of image retrieval and clustering algorithms depends on the effectiveness of the dissimilarity measure. ‘Minkowski’ distance, or more specifically, ‘Euclidean’ distance, is the most widely used dissimilarity measure in image retrieval and clustering. Euclidean distance depends only on the geometric position of two data instances in the feature space and completely ignores the data distribution. However, data distribution has an effect on human perception. The argument that two data instances in a dense area are more perceptually dissimilar than the same two instances in a sparser area, is proposed by psychologists. Based on this idea, a dissimilarity measure called, ‘mp’, has been proposed to address Euclidean distance’s limitation of ignoring the data distribution. Here, mp relies on data distribution to calculate the dissimilarity between two instances. As prescribed in mp, higher data mass between two data instances implies higher dissimilarity, and vice versa. mp relies only on data distribution and completely ignores the geometric distance in its calculations. In the aggregation of dissimilarities between two instances over all the dimensions in feature space, both Euclidean distance and mp give same priority to all the dimensions. This may result in a situation that the final dissimilarity between two data instances is determined by a few dimensions of feature vectors with relatively much higher values. As a result, the dissimilarity derived may not align well with human perception. The need to address the limitations of Minkowski distance measures, along with the importance of a dissimilarity measure that considers both geometric distance and the perceptual effect of data distribution in measuring dissimilarity between images motivated this thesis. It studies the performance of mp for image retrieval. It investigates a new dissimilarity measure that combines both Euclidean distance and data distribution. In addition to these, it studies the performance of such a dissimilarity measure for image retrieval and clustering. Our performance study of mp for image retrieval shows that relying only on data distribution to measure the dissimilarity results in some situations, where the mp’s measurement is contrary to human perception. This thesis introduces a new dissimilarity measure called, perceptual dissimilarity measure (PDM). PDM considers the perceptual effect of data distribution in combination with Euclidean distance. PDM has two variants, PDM1 and PDM2. PDM1 focuses on improving mp by weighting it using Euclidean distance in situations where mp may not retrieve accurate results. PDM2 considers the effect of data distribution on the perceived dissimilarity measured by Euclidean distance. PDM2 proposes a weighting system for Euclidean distance using a logarithmic transform of data mass. The proposed PDM variants have been used as alternatives to Euclidean distance and mp to improve the accuracy in image retrieval. Our results show that PDM2 has consistently performed the best, compared to Euclidean distance, mp and PDM1. PDM1’s performance was not consistent, although it has performed better than mp in all the experiments, but it could not outperform Euclidean distance in some cases. Following the promising results of PDM2 in image retrieval, we have studied its performance for image clustering. k-means is the most widely used clustering algorithm in scientific and industrial applications. k-medoids is the closest clustering algorithm to k-means. Unlike k-means which works only with Euclidean distance, k-medoids gives the option to choose the arbitrary dissimilarity measure. We have used Euclidean distance, mp and PDM2 as the dissimilarity measure in k-medoids and compared the results with k-means. Our clustering results show that PDM2 has perfromed overally the best. This confirms our retrieval results and identifies PDM2 as a suitable dissimilarity measure for image retrieval and clustering.
Doctor of Philosophy