A Design and Research of Image Detection System Based on Embedded

In recent years, terahertz imaging systems and techniques have been developed and have gradually become a leading frontier field. With the advantages of low radiation and clothing-penetrable, terahertz imaging technology has been widely used for the detection of concealed weapons or other contraband carried on personnel at airports and other secure locations. This paper aims to detect these concealed items with deep learning method for its well detection performance and real-time detection speed. Based on the analysis of the characteristics of terahertz images, an effective detection system is proposed in this paper. First, a lots of terahertz images are collected and labeled as the standard data format. Secondly, this paper establishes the terahertz classification dataset and proposes a classification method based on transfer learning. Then considering the special distribution of terahertz image, an improved faster region-based convolutional neural network (Faster R-CNN) method based on threshold segmentation is proposed for detecting human body and other objects independently. Finally, experimental results demonstrate the effectiveness and efficiency of the proposed method for terahertz image detection.

Download Full-text

Fabric Defect Detection System Using Stacked Convolutional Denoising Auto-Encoders Trained with Synthetic Defect Data

Applied Sciences ◽

10.3390/app10072511 ◽

2020 ◽

Vol 10 (7) ◽

pp. 2511

Author(s):

Young-Joo Han ◽

Ha-Jin Yu

Keyword(s):

Image Segmentation ◽

Deep Learning ◽

Machine Vision ◽

Image Classification ◽

Defect Detection ◽

Detection System ◽

Image Detection ◽

Fabric Defect Detection ◽

Fabric Defect ◽

Segmentation Image

As defect detection using machine vision is diversifying and expanding, approaches using deep learning are increasing. Recently, there have been much research for detecting and classifying defects using image segmentation, image detection, and image classification. These methods are effective but require a large number of actual defect data. However, it is very difficult to get a large amount of actual defect data in industrial areas. To overcome this problem, we propose a method for defect detection using stacked convolutional autoencoders. The autoencoders we proposed are trained by using only non-defect data and synthetic defect data generated by using the characteristics of defect based on the knowledge of the experts. A key advantage of our approach is that actual defect data is not required, and we verified that the performance is comparable to the systems trained using real defect data.

Download Full-text

An online identity authentication method for blood smear

Journal of Innovative Optical Health Sciences ◽

10.1142/s179354581550039x ◽

2016 ◽

Vol 09 (06) ◽

pp. 1550039

Author(s):

Xiaozhen Feng ◽

Yiping Cao ◽

Kuang Peng ◽

Cheng Chen

Keyword(s):

High Speed ◽

Blood Smear ◽

Detection System ◽

Identity Authentication ◽

Identification Accuracy ◽

Microscopic Image ◽

Online Identity ◽

Morphological Examination ◽

Medical Test ◽

Image Detection

Blood smear test is the basic method of blood cytology and is also a standard medical test that can help diagnose various conditions and diseases. Morphological examination is the gold standard to determine pathological changes in blood cell morphology. In the biology and medicine automation trend, blood smears' automated management and analysis is very necessary. An online blood smear automatic microscopic image detection system has been constructed. It includes an online blood smear automatic producing part and a blood smear automatic microscopic image detection part. Online identity authentication is at the core of the system. The identifiers printed online always present dot matrix digit code (DMDC) whose stroke is not continuous. Considering the particularities of DMDC and the complexities of online application environment, an online identity authentication method for blood smear with heterological theory is proposed. By synthesizing the certain regional features according to the heterological theory, high identification accuracy and high speed have been guaranteed with few features required. In the experiment, the sufficient correct matches between the tube barcode and the identification result verified its feasibility and validity.

Download Full-text

Design and simulation of the image detection system based on crack

2015 IEEE International Conference on Information and Automation ◽

10.1109/icinfa.2015.7279391 ◽

2015 ◽

Cited By ~ 1

Author(s):

Sen Wang ◽

Hong-jun San ◽

Xiao-qin Liu ◽

Xing Wu ◽

Tao Liu

Keyword(s):

Detection System ◽

Image Detection

Download Full-text

Image Detection System for Schmidt Plates

International Astronomical Union Colloquium ◽

10.1017/s0252921100106505 ◽

1984 ◽

Vol 78 ◽

pp. 169-171 ◽

Cited By ~ 1

Author(s):

Hideo Maehara ◽

Tomohiko Yamagata

Keyword(s):

Magnetic Tape ◽

Detection System ◽

Automated Detection ◽

The Other ◽

Image Detection ◽

Alternative Method ◽

Intermediate Medium

A 14-inch Schmidt plate contains 109 photographic grains and 105 to 106 images of stars and galaxies on it. Such a quantity of data is too large to be handled in a conventional way even for a big computer.There is, in general, an alternative method to solve this problem; one is to store the data of all pixels on intermediate medium (e.g., magnetic tape), and reduce them into image parameters afterwards. The other method is to do all the processing simultaneously with the measurement. The latter is very useful for the automated detection of celestial images on large Schmidt plates.

Download Full-text