# US 2014/0193071 A1 · Method and apparatus for detecting and recognizing object using local binary patterns

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2014/0193071 A1
- Document ID: US20140193071A1
- Status: Published application
- Kind: A1
- Application number: 13/899,003
- Filed: 2013-05-21
- Published: 2014-07-10
- Earliest claimed: 2013-01-10
- Assignees: Electronics and Telecommunications Research Institute
- Applicants: Electronics and Telecommunications Research Institute
- Inventors: Kyu Bo CHO; Yun Su CHUNG; Kil Taek LIM; Eun Chang CHOI; Soo In LEE
- Web page: https://patentgrep.com/patent/US20140193071A1

## CPC

- G06V 10/50: by performing operations within image blocks; by using histograms, e.g. histogram of oriented gradients [HoG]; by summing image-intensity values; Projection analysis
- G06V 20/52: Surveillance or monitoring of activities, e.g. for recognising suspicious objects (recognising microscopic objects G06V20/69)
- G06V 10/467: {Encoded features or binary features, e.g. local binary patterns [LBP]}
- G06V: IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING

## Abstract

A method and apparatus for detecting and recognizing an object using a vector histogram based on a local binary pattern are disclosed. The apparatus of detecting and recognizing an object using a local binary pattern includes: a feature map creator configured to extract an object area in which a moving object exists from an input image, to create a local binary pattern by designating a local area in the object area, and to create a vector component map including information about magnitude vector components and direction vector components using the local binary pattern; a feature map configuring unit configured to divide the object area into a plurality of blocks and to create a feature vector map through a histogram using the vector component map in a unit of the block; and an object detector configured to detect and classify the moving object based on the feature vector map.

## Claims

### Claim 1

An apparatus of detecting and recognizing an object using a local binary pattern, comprising:
a feature map creator configured to extract an object area in which a moving object exists from an input image, to create a local binary pattern by designating a local area in the object area, and to create a vector component map comprising information about magnitude vector components and direction vector components using the local binary pattern;
a feature map configuring unit configured to divide the object area into a plurality of blocks and to create a feature vector map through a histogram using the vector component map in a unit of the block; and
an object detector configured to detect and classify a moving object based on the feature vector map.

### Claim 2 (depends on 1)

The apparatus of claim 1, wherein the feature map creator comprises a local binary pattern creator configured to designate the local area in the object area, to compare the pixel value of the center pixel of the local area to the pixel values of adjacent pixels neighboring the center pixel to create binary pattern values, and to allocate the binary pattern values to the adjacent pixels, respectively.

### Claim 3 (depends on 2)

The apparatus of claim 2, wherein the local binary pattern creator applies predetermined weights to the binary pattern values to create a local binary pattern value and allocates the local binary pattern value to the center pixel to create the local binary pattern.

### Claim 4 (depends on 3)

The apparatus of claim 3, wherein the feature map creator further comprises a vector component map creator configured to add vectors whose start points are the center pixel and whose end points are the adjacent pixels based on the local binary pattern, thereby creating the vector component map.

### Claim 5 (depends on 1)

The apparatus of claim 1, wherein the feature map configuring unit comprises a feature point creator configured to accumulate the direction vector components weighted by the magnitude vector components on the histogram in a unit of the block, thereby creating feature points.

### Claim 6 (depends on 5)

The apparatus of claim 5, wherein the feature map configuring unit further comprises a feature vector normalizer configured to normalize feature vectors connecting the feature points to each other, thereby creating the feature vector map.

### Claim 7 (depends on 1)

The apparatus of claim 1, wherein the object detector detects and classifies the moving object comprising a person or a vehicle based on the feature vector map.

### Claim 8 (depends on 7)

The apparatus of claim 7, wherein the object detector classifies the moving object using a neural network method or a support vector machine (SVM).

### Claim 9 (depends on 1)

The apparatus of claim 1, wherein the input image is configured with a plurality of images photographed at predetermined time intervals.

### Claim 10

A method of detecting and recognizing an object using a local binary pattern, comprising:
extracting an object area in which a moving object exists from an input image, and creating a local binary pattern by designating a local area in the object area;
creating a vector component map comprising information about magnitude vector components and direction vector components using the local binary pattern; and
dividing the object area into a plurality of blocks, and creating a feature vector map through a histogram using the vector component map in a unit of the block.

### Claim 11 (depends on 10)

The method of claim 10, wherein the creating of the local binary pattern comprises designating the local area in the object area, comparing the pixel value of the center pixel of the local area to the pixel values of adjacent pixels neighboring the center pixel to create binary pattern values, and allocating the binary pattern values to the adjacent pixels, respectively.

### Claim 12 (depends on 11)

The method of claim 11, wherein the creating of the local binary pattern comprises applying predetermined weights to the binary pattern values to create a local binary pattern value and allocates the local binary pattern value to the center pixel to create the local binary pattern.

### Claim 13 (depends on 10)

The method of claim 10, wherein the creating of the vector component map comprises adding vectors whose start points are the center pixel and whose end points are the adjacent pixels based on the local binary pattern, thereby creating the vector component map.

### Claim 14 (depends on 10)

The method of claim 10, wherein the creating of the feature vector map comprises:
accumulating the direction vector components weighted by the magnitude vector components on the histogram in a unit of the block to create feature points; and
normalizing feature vectors connecting the feature points to each other to create the feature vector map.

### Claim 15 (depends on 10)

The method of claim 10, further comprising detecting and classifying the moving object based on the feature vector map.

### Claim 16 (depends on 15)

The method of claim 15, wherein the detecting and classifying of the moving object comprises detecting and classifying the moving object comprising a person or a vehicle using a neural network method or a support vector machine (SVM).

### Claim 17 (depends on 10)

The method of claim 10, wherein the input image is configured with a plurality of images photographed at predetermined time intervals.