Learning of fuzzy spatial relations between handwritten patterns

Poulenard

IEEE Trans. Pattern Anal. Mach. Intell.

et al. 2017

The analysis of spatial relations between objects in digital images plays a crucial role in various application domains related to pattern recognition and computer vision. Classical models for the evaluation of such relations are usually sufficient for the handling of simple objects, but can lead to ambiguous results in more complex situations. In this article, we investigate the modeling of spatial configurations where the objects can be imbricated in each other. We formalize this notion with the term enlacement, from which we also derive the term interlacement, denoting a mutual enlacement of two objects. Our main contribution is the proposition of new relative position descriptors designed to capture the enlacement and interlacement between two-dimensional objects. These descriptors take the form of circular histograms allowing to characterize spatial configurations with directional granularity, and they highlight useful invariance properties for typical image understanding applications. We also show how these descriptors can be used to evaluate different complex spatial relations, such as the surrounding of objects. Experimental results obtained in the different application domains of medical imaging, document image analysis and remote sensing, confirm the genericity of this approach.

Section: Related Workmentioning

confidence: 99%

Directional Enlacement Histograms for the Description of Complex Spatial Configurations between Objects

Poulenard

IEEE Trans. Pattern Anal. Mach. Intell.

et al. 2017

“…From these landscapes, fuzzy measures such as the necessity-possibility intervals [7] can be used to evaluate the relation for different target objects. Typical applications include for example graph-based face recognition [8], brain segmentation from MRI [9], or handwritten text recognition [10].…”

Section: Related Workmentioning

confidence: 99%

“…In the following, we show how to further exploit these landscapes to evaluate the degree to which a target object B is enlaced by A, using classical fuzzy operators such as necessity and possibility. Indeed, sev-eral works on spatial relations have used measures based on fuzzy sets to evaluate their approaches [8,9,10] (the reader can also refer to [32] for a summary of classical fuzzy measures). Let µ A and µ B be two fuzzy sets over R 2 .…”

Section: Fuzzy Evaluationsmentioning

confidence: 99%

Fuzzy directional enlacement landscapes for the evaluation of complex spatial relations

Wendling

2020

Pattern Recognition

Structural spatial relations between image components are fundamental in the human perception of image similarity, and constitute a challenging topic in the domain of image analysis. By definition, some specific relations are ambiguous and difficult to formalize precisely by humans. In this work, we deal with the issue of evaluating complex spatial configurations, where objects can surround each other, potentially with multiple levels of depth. Based on a recently introduced spatial relation called enlacement, which generalizes the idea of surrounding for arbitrary objects, we propose a fuzzy landscape model that allows both to visualize and evaluate this relation directly in the image space, following different directions. Experiments on several characteristic examples highlight the interest and the behavior of this approach, allowing for rich interpretations of these complex spatial configurations.

“…This approach is based on a fuzzy modeling of spatial relations directly in the image space, using morphological operations. Typical applications include for example graphbased face recognition [6], brain segmentation from MRI [8], or handwritten text recognition [12].…”

Section: Related Work a Spatial Relationsmentioning

confidence: 99%

Bags of spatial relations and shapes features for structural object description

2016 23rd International Conference on Pattern Recognition (ICPR)

Wendling

2016

We introduce a novel bags-of-features framework based on relative position descriptors, modeling both spatial relations and shape information between the pairwise structural subparts of objects. First, we propose a hierarchical approach for the decomposition of complex objects into structural subparts, as well as their description using the concept of Force Histogram Decomposition (FHD). Then, an original learning methodology is presented, in order to produce discriminative hierarchical spatial features for object classification tasks. The cornerstone is to build an homogeneous vocabulary of shapes and spatial configurations occurring across the objects at different scales of decomposition. An advantage of this learning procedure is its compatibility with traditional bags-of-features frameworks, allowing for hybrid representations of both structural and local features. Classification results obtained on two datasets of images highlight the interest of this approach based on hierarchical spatial relations descriptors to recognize structured objects.