National Taipei University of Technology

Institute for Information Industry

Chuan-Yen Chiang

Yen-Lin Chen

Chao-Wei Yu

Augustine Tsai

Meng-Tsan Li

Yang-Lang Chang

Wen-Yew Liang

A multi-class object classifying method and system are disclosed herein, where the multi-class object classifying method includes the following steps: classes, first training images and second training images are received and stored, and first characteristic images and second characteristic images are respectively extracted from the first training images and the second training images; the first training images is used to generate classifiers through a linear mapping classifying method; a classifier and the second characteristic images are used to determine parameter ranges corresponding to the classes and a threshold corresponding to the classifier. When two of the parameter ranges overlap, the remaining parameter ranges except for the two overlapped parameter ranges are recorded; after another classifier is selected from the classifiers except for the classifier that has been selected, the previous steps is repeated until the parameter ranges don't overlap with each other and the parameter ranges are recorded.

Multi-class object classifying method and system

The present invention discloses a multi-touch sensing system capable of optimizing touch bulbs according to the variation of ambient lighting conditions and a method thereof. The system comprises an image capturing module, a computing module and a processing module. The image capturing module captures a touch image. The computing module converts the touch image into a histogram and selects a grayscale threshold to segment the histogram by dichotomy for generating a segmented image of touch bulbs, and then calculates a between-class variance and a total pixel variance of the segmented image to estimate the separability factor thereof. The processing module determines whether or not the separability factor conforms to a predetermined value; if yes, then the processing module performs an image binarization of the touch image to generate a binary image, or else the processing module repeats the aforementioned process until the separability factor conforms to a predetermined value.

Multi-touch sensing system capable of optimizing touch bulbs according to variation of ambient lighting conditions and method thereof

The present invention is directed to a guide system having a function of a real-time voice response for the visually impaired and a method thereof. The guide system responds road conditions in real time using two processing modules performing image processing and voice responses, respectively. The guide system includes an visual sensing module sensing an image, a memory storing multiple training samples and multiple pieces of audio response information, an image processing module performing an image detection process to the image so as to create at least a segmented image, performing an object detection process to the segmented image, and performing an object recognition process so as to create a recognition signal, a system processing module creating an audio signal based on the recognition signal such that a speech voice is hearable by a user.

Guide system having function of real-time voice response for the visually impaired and method thereof

Growing community of inventors

Nantou County, Taiwan

Chuan-Yen Chiang