Binary code-based Human Detection
نویسندگان
چکیده
HOG features are effective for object detection, but their focus on local regions makes them highdimensional features. To reduce the memory required for the HOG features, this paper proposes a new feature, R-HOG, which creates binary codes from the HOG features extracted from two local regions. This approach enables the created binary codes to reflect the relationships between local regions. Converting feature values to binary, however, results in the loss of much information included in the features. In response to this problem, we have been focusing on“ quantization residual”information that is lost at this time. In this study, we introduce a transition likelihood model into the classifier based on two ideas using quantization residuals to consider the possibility that a binary code observed from the image will make a transition to another binary codes. This enables classification that takes into account all binary codes including the originally desired binary codes even if an observed binary code differs from the truly desired binary codes due to some sort of effect from another binary codes. Experimental results show that a classifier equipped with transition prediction based on quantization residuals as proposed here achieves high-accuracy human detection compared to the same classifier without transition prediction.
منابع مشابه
Algebraic Matching of Vulnerabilities in a Low-Level Code
This paper explores the algebraic matching approach for detection of vulnerabilities in binary codes. The algebraic programming system is used for implementing this method. It is anticipated that models of vulnerabilities and programs to be verified are presented as behavior algebra and action language specifications. The methods of algebraic matching are based on rewriting rules and techniques...
متن کاملDynamic Malicious Code Detection Based on Binary Translator
The binary translator is a software component of a computer system. It converts binary code of one ISA into binary code of another ISA. Recent trends show that binary translators have been used to save CPU power consumption and CPU die size, which makes binary translators a possible indispensable component of future computer systems. And such situation would give new opportunities to the securi...
متن کاملBitShred: Fast, Scalable Code Reuse Detection in Binary Code
Many experts believe that new malware is created at a rate faster than legitimate software. For example, in 2007 over one million new malware samples were collected by a major security solution vendor. However, it is often speculated, though to the best of our knowledge unproven, that new malware is produced by modifying existing malware, either through simple tweaks, code composition, or a var...
متن کاملBitShred: Fast, Scalable Code Reuse Detection in Binary Code (CMU-CyLab-10-006)
Many experts believe that new malware is created at a rate faster than legitimate software. For example, in 2007 over one million new malware samples were collected by a major security solution vendor. However, it is often speculated, though to the best of our knowledge unproven, that new malware is produced by modifying existing malware, either through simple tweaks, code composition, or a var...
متن کاملStatic Detection of Malicious Code in Executable Programs
In this paper, we propose a new approach for the static detection of malicious code in executable programs. Our approach rests on a semantic analysis based on behaviour that even makes possible the detection of unknown malicious code. This analysis is carried out directly on binary code. Static analysis offers techniques for predicting properties of the behaviour of programs without running the...
متن کامل