Combining image regions and human activity for indirect object recognition in indoor wide-angle views

Traditional methods of object recognition are reliant on shape and so are very difficult to apply in cluttered, wideangle and low-detail views such as surveillance scenes. To address this, a method of indirect object recognition is proposed, where human activity is used to infer both the location and identity of objects. No shape analysis is necessary. The concept is dubbed 'interaction signatures', since the premise is that a human will interact with objects in ways characteristic of the function of that object - for example, a person sits in a chair and drinks from a cup. The human-centred approach means that recognition is possible in low-detail views and is largely invariant to the shape of objects within the same functional class. This paper implements a Bayesian network for classifying region patches with object labels, building upon our previous work in automatically segmenting and recognising a human's interactions with the objects. Experiments show that interaction signatures can successfully find and label objects in low-detail views and are equally effective at recognising test objects that differ markedly in appearance from the training objects.

Determining motion components using the point distribution model

The Point Distribution Model (PDM) has proven effective in modelling variations in shape in sets of images, including those in which motion is involved such as body and hand tracking. This paper proposes an extension to the PDM through a re-parameterisation of the model which uses factors such as the angular velocity and distance travelled for sets of points on a moving shape. This then enables non-linear quantities such as acceleration and the average velocity of the body to be expressed in a linear model by the PDM. Results are shown for objects with known acceleration and deceleration components, these being a simulated pendulum modelled using simple harmonic motion and video sequences of a real pendulum in motion.

Analysis of spherical indentation of superelastic shape memory alloys

Dimensional analysis and the finite element method are applied in this paper to study spherical indentation of superelastic shape memory alloys. The scaling relationships derived from dimensional analysis bridge the indentation response and the mechanical properties of a superelastic shape memory alloy. Several key variables of a superelastic indentation curve are revealed and examined. We prove that the bifurcation force in a superelastic indentation curve only relies on the forward transformation stress and the elastic properties of the initial austenite; and the return force in a superelastic indentation curve only relies on the reverse transformation stress and the elastic properties of the initial austenite. Furthermore, the dimensionless functions to determine the bifurcation force and the return force are proved to be identical. These results not only enhance our understanding of spherical indentation of superelastic shape memory alloys, but also provide the theoretical basis for developing a practicable method to calibrate the mechanical properties of a superelastic material from the spherical indentation test.

Spherical indentation hardness of shape memory alloys

Using dimensional analysis and the finite element method, the spherical indentation hardness of shape memory alloys (SMAs) is investigated. The scaling relationship between the hardness and the mechanical properties of a SMA, such as the forward transformation stress, the maximum transformation strain magnitude, has been derived. Numerical results demonstrated that the hardness increases with the indentation depth but there is no three-fold relationship between the hardness and the forward transformation stress. Increasing the maximum transformation strain magnitude would reduce the hardness of the material. These research results enhance our understanding of the hardness from the spherical indentation of SMAs.

Analysis of stamping production data with view towards quality management

A quality analysis trial was undertaken at Ford Geelong Stamping Plant on a press line that was fitted with standard press sensors to measure press and binder force over the stamping cycle for each panel. The quality of randomly sampled panels was measured by obtaining the panel thicknesses at five points, for 135 panels in total. These points were chosen such that they exhibited different forming modes. This paper analyses the input force data and the output quality data from the trial to determine any potential relationships. The analysis of the production data was performed using statistical correlation techniques to determine initial potential relationships between input and output variables. An Active Shape Model was used to extract features when identifying the major sources of variation within the input data. However, the initial analysis of the data elicited no direct relationship between the input variables measured and the panel thicknesses. This result is significant as the data collected is from a standard sensor configuration found in many press lines through-out the world. The reason for the lack of a direct relationship is believed to come from the lack of sensitivity in the force measurements which are not able to identify small changes in the process, whereas gross geometric variations have in previous studies shown an obvious relationship with changes in the force press profile. This means that existing force sensors require augmentation by additional sensors if a detailed automatic quality control system for the press lines based on input sensors alone.

Multi-scale analysis of connectivity for image retrieval

Previously, we proposed the concept of connectivity to obtain discriminating shape descriptors. In this paper, we use connectivity to obtain superior distance histograms for multi-scale images. Experiments are performed to evaluate the distance histograms, based on connectivity, for shape-based retrieval of multi-scale images. Item S8 within the MPEG-7 still images content set is used for performing experiments. Experimental results show that the proposed method enhances retrieval performance significantly.

TiNi shape memory alloy foams synthesized by spacer sintering and their properties

Titanium (Ti) and nickel (Ni) elemental powders were blened by ball milling and the ball milled powders were employed to fabricate NiNi shape memory alloy (SMA) foams by space sintering. Effect of ball milling time on phase constitutes of the sintered TiNi alloy foams was studied by X-ray diffraction (XRD) analysis.Scanning election microscopy (SEM) was used to characterize the porous structure, and compressive tests were carried out to evaluate the mechanical properties of the foams. Results indicate the porosities of the TiNi alloy foams can be controlled by using the spacer sincering method, and the porosities show a significant effect on the mechanical prperties and shape memory effect (SME).

Shape preserving approximation using least squares splines

Least squares polynomial splines are an effective tool for data fitting, but they may fail to preserve essential properties of the underlying function, such as monotonicity or convexity. The shape restrictions are translated into linear inequality conditions on spline coefficients. The basis functions are selected in such a way that these conditions take a simple form, and the problem becomes non-negative least squares problem, for which effecitive and robust methods of solution exist. Multidimensional monotone approximation is achieved by using tensor-product splines with the appropriate restrictions. Additional inter polation conditions can also be introduced. The conversion formulas to traditional B-spline representation are provided.

Titanium–nickel shape memory alloy foams for bone tissue engineering

Titanium–nickel (TiNi) shape memory alloy (SMA) foams with an open-cell porous structure were fabricated by space-holder sintering process and characterized by scanning electron microscopy (SEM) and X-ray diffraction (XRD) analysis. The mechanical properties and shape memory properties of the TiNi foam samples were investigated using compressive test. Results indicate that the plateau stresses and elastic moduli of the foams under compression decrease with the increase of their porosities. The plateau stresses and elastic moduli are measured to be from 1.9 to 38.3 MPa and from 30 to 860 MPa for the TiNi foam samples with porosities ranged from 71% to 87%, respectively. The mechanical properties of the TiNi alloy foams can be tailored to match those of bone. The TiNi alloy foams exhibit shape memory effect (SME), and it is found that the recoverable strain due to SME decreases with the increase of foam porosity.

Identification of cashmere and superfine merino wool with wavelet texture analysis

This paper presents the use of the wavelet transform to extract fibre surface texture features for classifying cashmere and superfine merino wool fibres. To extract features from brightness variations caused by the cuticular scale height, shape and interval provides an effective way for characterising different animal fibres and subsequently classifying them. This may enable the development of a completely automated and objective system for animal fibre
identification.

Process analysis of the solution strategies used for problems contained in the Minnesota Paper Board Form Board test

This thesis reviews progress toward an understanding of the processes involved in the solution of spatial problems. Previous work employing factor analysis and information processing analysis is reviewed and the emphasis on variations in speed and accuracy as the major contributers to individual differences is noted. It is argued that the strategy used by individuals is a preferable explanatory concept for identifying the cognitive substratum necessary for problem solving. Using the protocols obtained from subjects solving The Minnesota Paper Form Board (Revised), a test commonly regarded as measuring skill in spatial visualization, a number of different strategies are isolated. Assumptions as to the task variants which undergird these strategies are made and tested experimentally. The results suggest that task variants such as the size of the stimulus and the shape of the pieces interact with subject variables to produce the operating strategy. Skill in problem solving is revealed in the ability to structure the array, to hold a structured image and to reduce the number of answers requiring intensive processing. The interaction between task and subject variables results in appropriate or inappropriate strategies which in turn affect speed and accuracy. Results suggest that strategy formation and usage are the keys to explaining individual differences and an heuristic model is presented to explain the performance of individual subjects on the problems involved in the Minnesota Paper Form Board. The model can be used to predict performance on other tests; and as an aid to teaching subjects experiencing difficulties. The model presented incorporates strategy variation and is consequently mores complex than previously suggested models. It is argued that such complexity is necessary to explain the nature of a subject's performance and is also necessary to perform diagnostic evaluation. Certain structural -features of the Minnesota Paper Form Board are questioned and suggestions for improvement included. The essential explanatory function of the strategy in use makes the prevalent group administration approach suspect in the prediction of future performance in spatial or vocational activity.

Analysis of springback variation in sheet metal forming

This thesis investigated how the variation in inputs, such as material and processing conditions affected the shape defect phenomenon (springback) for sheet metal forming processes. Using a stochastic Finite Element modelling tool, it was found that the material type and fluctuations in material properties significantly influenced the variation in springback.

Haptics enabled offline AFM image analysis

Current advancements in nanotechnology are dependent on the capabilities that can enable nano-scientists to extend their eyes and hands into the nano-world. For this purpose, a haptics (devices capable of recreating tactile or force sensations) based system for AFM (Atomic Force Microscope) is proposed. The system enables the nano-scientists to touch and feel the sample surfaces, viewed through AFM, in order to provide them with better understanding of the physical properties of the surface, such as roughness, stiffness and shape of molecular architecture. At this stage, the proposed work uses of ine images produced using AFM and perform image analysis to create virtual surfaces suitable for haptics force analysis. The research work is in the process of extension from of ine to online process where interaction will be done directly on the material surface for realistic analysis.