J. Martínez-Trinidad

The logical combinatorial approach to pattern recognition, an overview through selected works

Abstract The so-called logical combinatorial approach to Pattern Recognition is presented, and works (mainly in Spanish and Russian) that are not ordinarily available, are exposed to the Western reader. The use of this approach for supervised and unsupervised pattern recognition, and for feature selection is reviewed. Also, an unified notation describing the original contributions is presented, thus rendering this important area more readable. Our review is not exhaustive; nevertheless, most significant works are enclosed. Our hope is to motivate the reader to inquire further in these works. This paper serves as an introduction to three articles on the logical combinatorial approach that appear in this issue of Pattern Recognition.

Characterisation of CoB–Co2B coatings by the scratch test

New results about the scratch practical adhesion-resistance of the CoB–Co2B/substrate system developed at the surface of CoCrMo (ASTM-F75) alloy were estimated. The boron diffusion on the surface of the cobalt alloy was conducted using the powder-pack boriding process at temperatures of 1223 and 1273 K with different exposure times for each temperature. The scratch tests over the surface of cobalt borided alloy were performed with a 200 micrometres Rockwell C diamond indenter considering a continuously increasing normal force for the entire set of experimental conditions of the boriding process. The worn tracks produced on the coating/substrate system were analysed by optical and scanning electron microscopy to estimate and identify the critical loads and failure mechanisms, respectively. The results indicated that the critical loads varied between 95 and 142 N as a function of the boride coating thicknesses with a development of various types of failure mechanisms over the surface of the coating/substrate system.

Properties and Characterization of Hard Coatings Obtained by Boriding: An Overview

Some physicochemical and mechanical properties of surface hard coatings obtained by the paste-boriding process are summarized in this work. Different grades of borided ferrous alloys were used to develop the formation of surface layers type Fe2B or FeB/Fe2B. Furthermore, in order to characterize the nature of boride layers, some classical techniques are presented and discussed such as Glow Discharge Optical Emission Spectrometry (GDOES), Atomic Force Microscopy (AFM) and estimation of residual stresses by X-Ray Diffraction method. Also, the morphology of borided interfaces was evaluated by concepts of fractal theory.

Evaluation of Brittle Layers Obtained by Boriding on AISI H13 Steels

The fracture toughness of AISI H13 borided steel and the strength adhesion of the coated system were estimated in the present work. The formation of the layers was carried out by the powder pack boriding process at 1273 K with 8 h of treatment. The fracture toughness (KC) of the layer is estimated at 25 and 45 m from the surface using four different Vickers indentation loads. The KC values were estimated by the extension of Palmqvist cracks parallel and perpendicular to the surface obtained at the indentation corners. The adherence of the layer/substrate was evaluated in qualitative form through the Rockwell-C indentation technique. The results obtained by both techniques, show, in first instance, that the fracture toughness of the boride layer can be expressed in the form (KC) (π/2) > (KC) > (KC) (0). Also, high delamination is obtained around the Rockwell-C indentation prints that denote poor adhesion in the coating-substrate interface.

Fuzzy clustering of semantic spaces

Abstract In this paper the GK + model for the construction of thesaurus classes based on fuzzy semantic association measure between index terms and concepts (thesaurus classes) is presented. The association measure is obtained on the basis of fuzzy semantic relations between index terms, and it is used to cluster index terms into concepts. A hierarchical algorithm is introduced which runs on a simple numerical example.

Dependence between the Boron Surface Concentration and the Growth Kinetics of Boride Layers in AISI 4140 Steels

The present work estimated the growth kinetics of Fe2B layers formed at the surface of AISI 4140 steels. The thermochemical treatment was applied in order to produce the Fe2B phase, considering temperatures of 1123, 1173, 1223 and 1273 K with five exposure times (2, 4, 5, 6, and 8 h), using a 4 mm thick layer of boron carbide paste over the material surface. The growth of boride layers was described by the mass balance equation between phases in thermodynamic equilibrium, assuming that the growth of boride layers obeys the parabolic growth equation and the boron concentration at the interfaces remains constant. Also, the boron diffusion coefficient at the Fe2B ( ) was established as a function of boriding temperature. Likewise, the parabolic growth constant (k), the instantaneous velocity (v) of the Fe2B/substrate interface and the weight-gain of borided steels were established as a function of the parameters and , which are related to the boride incubation time ( ) and boron surface concentration ( ), respectively.

Anisotropy of Boride Layers: Effect on the Mechanical Properties of AISI 4140 Borided Steels

The growth of iron borides over the surface of different steels is of high anisotropy. It was determined that the anisotropy of borided phases reveals a significant instability of properties in service. One of the techniques to determine the effect of anisotropy on the mechanical properties of boride layers is the induced-fracture by Vickers microindentation. During the present work, the fracture toughness (KC) of the Fe2B hard coatings has been estimated at the surface of AISI 4140 borided steels. The force criterion of fracture toughness was determined from the extent of brittle cracks originating at the tips of an indenter impression. The indentation loads were established between 1.9 to 9.8 N at three different distances from the borided surface. The KC values were expressed as a function of temperature, treatment time and the indentation distances from the surface. Likewise, the adherence of the coated system was evaluated by Rockwell-C indentation, where the borided steel showed sufficient adhesion.

Effect of Boriding on the Mechanical Properties of AISI 1045 Steel

Some mechanical properties of AISI 1045 borided steels were estimated in the present work. The boriding process was carried out by the powder pack method at 950°C with 8 h of treatment. The fatigue strength on borided notched specimens was evaluated with rotating bending tests (R=1) considering a stress concentration factor (Kt) of 2.53. Likewise, the presence of residual stresses in boride layers was established by the XRD technique. The Daimler-Benz Rockwell C test was used, also, to estimate the strength adhesion of the coated system. The results show a decrease in the fatigue strength of AISI borided steels due to the presence of high porosity in the layers. Finally, the Rockwell-C adhesion test showed no coating failure for the boride layer.

Fracture Indentation on AISI 1018 Borided Steels

Fracture indentation was applied to estimate the fracture toughness of AISI 1018 borided steels. The Fe2B hard layers were formed using the powder-pack boriding process for two temperatures with 4 and 8 h of exposure times. The fracture toughness of the iron boride layer of the AISI 1018 borided steels was estimated using a Vickers microindentation induced-fracture testing at distances of 15 and 30 m from the surface, applying four loads (0.49, 0.98, 1.96 and 2.9 N). The microcracks generated at the corners of the Vickers microindentation were considered as experimental parameters, which are introduced in a Palmqvist crack model to determine their corresponding fracture toughness KC. As a result, the experimental parameters, such as exposure time and boriding temperature are compared with the resulting fracture toughness of the borided phase.

Multi-graph Frequent Approximate Subgraph Mining for Image Clustering

In data mining, frequent approximate subgraph (FAS) mining techniques has taken the full attention of several applications, where some approximations are allowed between graphs for identifying important patterns. In the last four years, the application of FAS mining algorithms over multi-graphs has reported relevant results in different pattern recognition tasks like supervised classification and object identification. However, to the best of our knowledge, there is no reported work where the patterns identified by a FAS mining algorithm over multi-graph collections are used for image clustering. Thus, in this paper, we explore the use of multi-graph FASs for image clustering. Some experiments are performed over image collections for showing that by using multi-graph FASs under the bag of features image approach, the image clustering results reported by using simple-graph FAS can be improved.