Real Time Machine Learning Based Car Detection in Images With Fast Training

Our primary interest is to build fast and reliable object recognizers in images based on small training sets. This is important in cases where the training set needs to be built mostly manually, as in the case that we studied, the recognition of the Honda Accord 2004 from rear views. We describe a novel variant of the AdaBoost based learning algorithm, which builds a strong classifier by incremental addition of weak classifiers (WCs) that minimize the combined error of the already selected WCs. Each WC is trained only once, and examples do not change their weights. We describe a set of appropriate feature types for the considered recognition problem, including a redness measure and dominant edge orientations. The existing edge orientation bin division was improved by shifting so that all horizontal (vertical, respectively) edges belong to the same bin. We propose to pre-eliminate features whose best threshold value is near the trivial position at the minimum or maximum of all threshold values. This is a novel method that has reduced the training set WC quantity to less than 10% of its original number, greatly speeding up training time, and showing no negative impact on the quality of the final classifier. We also modified the AdaBoost based learning machine. Our experiments indicate that the set of features used by Viola and Jones and others for face recognition was inefficient for our problem, recognizing cars accurately and in real time with fast training. Our training method has resulted in finding a very accurate classifier containing merely 30 WCs after about 1 h of training. Compared to existing literature, we have overall achieved the design of a real time object detection machine with the least number of examples, the least number of WCs, the fastest training time, and with competitive detection and false positive rates.     

Structure and properties of continuously cast low-carbon steel as a function of the reeling temperature

Investigation of continuously cast low-carbon aluminum-killed steel has great practical importance. Its structure and properties depend on the reeling temperature after hot rolling. After reeling at a low temperature the steel exhibits a capacity for deep drawing, whereas after reeling at a high temperature it is suitable for enameling.Translated from Metallovedenie i Tetmicheskaya Obrabotka Metallov, No. 8, pp. 23 – 24, August, 1996.     

Material flow and microstructure in the friction stir butt welds of the same and dissimilar aluminum alloys

The material flow and microstructural evolution in the friction stir welds of a 6061-Al alloy to itself and of a 6061-Al alloy to 2024-Al alloy plates of 12.7 mm in thickness were studied under different welding conditions. The results showed that plastic deformation, flow, and mechanical mixing of the material exhibit distinct asymmetry characteristics at both sides of the same and dissimilar welds. The microstructure in dissimilar 6061-Al/2024-Al welds is significantly different from that in the welds of a 6061-Al alloy to itself. Vortex-like structures featured by the concentric flow lines for a weld of 6061-Al alloy to itself, and alternative lamellae with different alloy constituents for a weld of 6061-Al to 2024-Al alloy, are attributed to the stirring action of the threaded tool, in situ extrusion, and traverse motion along the welding direction. The mutual mixing in the dissimilar metal welds is intimate and far from complete. However, the bonding between the two Al-alloys is clearly complete. Three different regions in the nugget zone of dissimilar 6061-Al/2024-Al welds are classified by the mechanically mixed region (MMR) characterized by the relatively dispersed particles of different alloy constituents, the stirring-induced plastic flow region (SPFR) consisting of alternative vortex-like lamellae of the two Al-alloys, and the unmixed region (UMR) consisting of fine equiaxed grains of the 6061-Al alloy. Within all of these three regions, the material is able to withstand a very high degree of plastic deformation due to the presence of dynamic recovery or recrystallization of the microstructure. The degree of material mixing, the thickness of the deformed Al-alloy lamellae, and the material flow patterns depend on the related positions in the nugget zone and the processing parameters. Distinct fluctuations of hardness are found to correspond to the microstructural changes throughout the nugget zone of dissimilar welds.     

Understanding the Influence of Copper Nanoparticles on Thermal Characteristics and Microstructural Development of a Tin-Silver Solder

This paper presents and discusses issues relevant to solidification of a chosen lead-free solder, the eutectic Sn-3.5%Ag, and its composite counterparts. Direct temperature recordings for the no-clean solder paste during the simulated reflow process revealed a significant amount of undercooling to occur prior to the initiation of solidification of the eutectic Sn-3.5%Ag solder, which is 6.5 °C, and for the composite counterparts, it is dependent on the percentage of copper nanopowder. Temperature recordings revealed the same temperature level of 221 °C for both melting (from solid to liquid) and final solidification (after recalescence) of the Sn-3.5%Ag solder. Addition of copper nanoparticles was observed to have no appreciable influence on melting temperature of the composite solder. However, it does influence solidification of the composite solder. The addition of 0.5 wt.% copper nanoparticles lowered the solidification temperature to 219.5 °C, while addition of 1.0 wt.% copper nanoparticles lowered the solidification temperature to 217.5 °C, which is close to the melting point of the ternary eutectic Sn-Ag-Cu solder alloy, Sn-3.7Ag-0.9Cu. This indicates the copper nanoparticles are completely dissolved in the eutectic Sn-3.5%Ag solder and precipitate as the Cu₆Sn₅, which reinforces the eutectic solder. Optical microscopy observations revealed the addition of 1.0 wt.% of copper nanoparticles to the Sn-3.5%Ag solder results in the formation and presence of the intermetallic compound Cu₆Sn_5. These particles are polygonal in morphology and dispersed randomly through the solder matrix. Addition of microsized copper particles cannot completely dissolve in the eutectic solder and projects a sunflower morphology with the solid copper particle surrounded by the Cu₆Sn₅ intermetallic compound coupled with residual porosity present in the solder sample. Microhardness measurements revealed the addition of copper nanopowder to the eutectic Sn-3.5%Ag solder resulted in higher hardness.     

Sensing, modeling and control for laser-based additive manufacturing

Laser-Based Additive Manufacturing (LBAM) is a promising manufacturing technology that can be widely applied to part preparation, surface modification, and Solid Freeform Fabrication (SFF). A large number of parameters govern the LBAM process. These parameters are sensitive to the environmental variations, and they also influence each other. This paper introduces the research work in RCAM on improving the performance of the LBAM process. Metal powder delivery real-time sensing and control is studied to achieve a controllable powder delivery for fabrication of functionally graded material. A closed-loop control system based on infrared image sensing is built for control of the heat input and size of the molten pool in the LBAM process. The closed-loop control results show a great improvement in the geometrical accuracy of the built features. A three-dimensional finite element model is also established to explore the thermal behavior of the molten pool in the closed-loop controlled LBAM process.     

On the role of hydrophilicity and hydrophobicity in aqueous heterophase polymerization

This paper reports on experimental results of aqueous heterophase polymerizations with monomers of quite different solubility in water ranging from the water-soluble 2-hydroxyethyl methacrylate to lauryl methacrylate with solubility in water of only about 10⁻⁴ mM. A calorimetric study revealed the strong influence of both the hydrophilicity of the monomer and the stirrer speed on the rate of polymerization in the absence of surfactants. In order to obtain maximum latex yield and high efficiency (which is a measure considering colloidal properties, polymerization recipe, and polymerization parameters) the initiator-surfactant combination must be properly chosen in dependence on the hydrophilicity of the monomer. Results are presented for sodium alkyl sulfates or disodium-N-stearoyl-l-glutamate as surfactant and potassium peroxodisulfate, or poly(ethylene glycol)-azo- compounds, or 2,2′-azobis(N-2′-methylpropanoyl-2-amino-alkyl-1)-sulfonates as initiators.     

An alignment-free fingerprint bio-cryptosystem based on modified Voronoi neighbor structures

Bio-cryptography is an emerging security technology which combines cryptography with biometrics. A good bio-cryptosystem is required to protect the privacy of the relevant biometric data as well as achieving high recognition accuracy. Fingerprints have been widely used in bio-cryptosystem design. However, fingerprint uncertainty caused by distortion and rotation during the image capturing process makes it difficult to achieve a high recognition rate in most bio-cryptographic systems. Moreover, most existing bio-cryptosystems rely on the accurate detection of singular points for fingerprint image pre-alignment, which is very hard to achieve, and the image rotation transformation during the alignment process can cause significant singular point deviation and minutiae changes. In this paper, by taking full advantage of local Voronoi neighbor structures (VNSs), e.g. local structural stability and distortion insensitivity, we propose an alignment-free bio-cryptosystem based on fixed-length bit-string representations extracted from modified VNSs, which are rotation- and translation-invariant and distortion robust. The proposed alignment-free bio-cryptosystem is able to provide strong security while achieving good recognition performance. Experimental results in comparison with most existing alignment-free bio-cryptosystems using the publicly-available databases show the validity of the proposed scheme.     

Automatic player position detection in basketball games

This paper presents us with a framework for the automatic player position detection (APPD) in the game of basketball. Court players are detected in the images broadcasted via television stations. In them, at any point of time, the view is from only one camera. This makes the detection process much more difficult. The player detection is based on the mixture of non-oriented pictorial structures. The detection of body parts is performed by the Support Vector Machine (SVM) algorithm. The results of these detections are combined together with constraints on their locations, which specify the position of one body part with respect to the parent body part. In order to train the whole model, we used a latent form of SVM called the latent SVM (LSVM). Such approach generated the statistical accuracy of about 82 %, which represents one of the best results in basketball player detection framework. Beside players, the algorithm detected a certain number of false positive objects. These are mostly people from the audience and the referees as well. This paper contains a simple and robust solution to remove them all, based on the play court boundaries and the histogram comparison. Separating players in different teams is done by k-means clustering. The inputs to this algorithm are saturation histograms calculated on the jerseys. A spatial transformation is determined by the detected play court boundaries and the actual court measures. Using this transformation, points representing the location of detected players in TV images are mapped to the actual location of players on the court, which was the main goal of our research. The proposed solution is sound and efficient. In addition, it is backed up by the experimental results obtained using the model of the actual footage of basketball games.     

Modeling of waterjet guided laser grooving of silicon

Waterjet guided laser processing is an internationally patented technique based on guiding a laser inside a thin, high-speed waterjet. The process combines the advantages of laser processing with those of waterjet cutting and offers promise as a method for processing thin and heat sensitive materials with a high degree of precision. An improved understanding of the complex interaction between laser, waterjet, and workpiece is required to enable the process to achieve its potential. A model for waterjet guided laser grooving of silicon is presented that treats the energy input of the laser, the cooling effect of the waterjet, and the melting and removal of the silicon. The thermal process is represented in detail in the new method. The model is validated by comparisons of simulation and experimental results, and the simulation provides insight regarding the details of the interactions among laser, waterjet, and workpiece.     

Subband coding systems incorporating quantizer models

A new method for dealing with the effects of quantization in a subband system is proposed. It uses the "gain plus additive noise" linear model for the Lloyd-Max quantizer. Based on this, it is demonstrated how, by an appropriate choice of synthesis filters, one can cancel all signal-dependent errors at the output of the system. The only remaining error is random in nature and not correlated with the input signal. We therefore have a tradeoff between the error being only random or having signal-dependent components as well (since the error variances in both cases are comparable). As a result of having only a random error, it is possible to reduce this error using, for example, a noise removal technique. The result is then extended to the case where the input is a multidimensional signal, and arbitrary sampling lattices are used, as well as to the QMF (alias cancellation) case. To demonstrate the validity of the proposed approach, two types of experiments on images are carried out: In a toy example, it is shown that using noise removal could be beneficial. For a more realistic coding scheme, however, it is demonstrated that even in the case when the model is no longer valid (when some of the subbands are discarded), the output error is still much less correlated with the input signal as opposed to the commonly used subband system, while visually, the reconstructed images look very similar.