Improved learning of I2C distance and accelerating the neighborhood search for image classification

Image-to-class (I2C) distance is a novel measure for image classification and has successfully handled datasets with large intra-class variances. However, due to the lack of a training phase, the performance of this distance is easily affected by irrelevant local features that may hurt the classification accuracy. Besides, the success of this I2C distance relies heavily on the large number of local features in the training set, which requires expensive computation cost for classifying test images. On the other hand, if there are small number of local features in the training set, it may result in poor performance.In this paper, we propose a distance learning method to improve the classification accuracy of this I2C distance as well as two strategies for accelerating its NN search. We first propose a large margin optimization framework to learn the I2C distance function, which is modeled as a weighted combination of the distance from every local feature in an image to its nearest-neighbor (NN) in a candidate class. We learn these weights associated with local features in the training set by constraining the optimization such that the I2C distance from image to its belonging class should be less than that to any other class. We evaluate the proposed method on several publicly available image datasets and show that the performance of I2C distance for classification can significantly be improved by learning a weighted I2C distance function. To improve the computation cost, we also propose two methods based on spatial division and hubness score to accelerate the NN search, which is able to largely reduce the on-line testing time while still preserving or even achieving a better classification accuracy.     

An information-leak analysis system based on program slicing

For programs using secret information such as credit card numbers, preventing information-leaks is important. Denning, for example, has proposed a mechanism to certify that a given program does not violate a security policy. Kuninobu, on the other hand, has proposed a more practical framework for calculating the secrecy level of each output value from the secrecy level set to each input value, but no implementation has been yet explored. In this paper, we propose an implementation method for information-leak analysis, and show a system we have implemented based on program slicing. We have applied this system to a credit card program. Our results show that information-leak analysis before practical use of the program is important.     

Characterization of polypropylene–polyethylene blends made of waste materials with compatibilizer and nano-filler

Waste polypropylene and polyethylene were blended by a twin-screw extruder with two compatibilizers (PE-g-MAH and EPDM) and an additive (O-MMT). The mechanical properties were measured firstly. By adding O-MMT, the tensile strength showed a decline while the impact strength made a promotion. The phase morphology was observed by scanning electron microscopy (SEM) to explore the fracture toughness of blends. The blend with EPDM had a better compatibilization than PE-g-MAH. X-ray diffraction was used to investigate the crystallization behavior and the result showed no change by blending. Moreover, further measurements such as thermogravimetric (TGA) and differential scanning calorimetry (DSC) were taken to show the thermal stability and crystallization temperature of the blend. Additionally, the storage modulus and loss modulus are measured by dynamic mechanical analysis (DMA), the presence of O-MMT caused the increases of the storage modulus and loss modulus.     

Microstructure and mechanical properties of laser beam welded Al–Li alloy 2060 with Al–Mg filler wire

Alloy 2060-T8 is a newly developed high-strength Al–Li alloy for applications in aircraft industry. Crack-free welds were obtained in laser beam welding with 5087 filler wire under optimized welding conditions. In this paper, fusion zone microstructure and joint mechanical properties were investigated. Microstructure typical for the weld metal consists of α-Al matrix with a few nanoscale precipitates inside and a coarse icosahedral quasicrystalline T₂ phase at the dendritic and grain boundaries. The quasicrystalline occurred normally in Al–Li–Cu alloys with higher Li contents. Our investigations show that the icosahedral quasicrystalline phase T₂ phase forms in the laser-welded Al–Li alloy 2060 with lower Li content as a result of segregation and replacement of Mg element. The joint tensile strength in as-welded condition is around 317 MPa, about 63% of that of the base metal, and fracture occurs within the fusion zone.