Industrial Technology Research Institute

Fuh-Yu Chang

Min-Chieh Chou

Kai-Tai Song

Hung-Yi Lin

Wen-Yang Peng

Meng-Ju Han

Jiu-Shu Tsai

Ching-Hsiang Cheng

Shao-Heng Chang

Jing-Huai Hsu

Jung-Wei Hong

A method is disclosed in the present disclosure for recognizing emotion by setting different weights to at least of two kinds of unknown information, such as image and audio information, based on their recognition reliability respectively. The weights are determined by the distance between test data and hyperplane and the standard deviation of training data and normalized by the mean distance between training data and hyperplane, representing the classification reliability of different information. The method recognizes the emotion according to the unidentified information having higher weights while the at least two kinds of unidentified information have different result classified by the hyperplane and correcting wrong classification result of the other unidentified information so as to raise the accuracy while emotion recognition. Meanwhile, the present disclosure also provides a learning step with a characteristic of higher learning speed through an algorithm of iteration.

Bimodal emotion recognition method and system utilizing a support vector machine

The present invention relates to a method of fabricating a vertical probe head, whereas the vertical probe head is formed by the combination of at least a probe, a bottom guide plate and a top guide plate having at least a hole matching the probe. The probe is fabricated by a LIGA-like process combining with the processes of photolithography, etching and electroforming, and so on, so that the probe is equipped with comparatively better precision, strength and reliability and yet can be custom-made for satisfying various demands. In addition, both the top and bottom guide plates are made by a means of non-mechanical machining, which respectively is fabricated by processing a substrate using means of photolithography, etching and mask so as to fabricate holes for matching with the aforesaid probe. The vertical probe head of the invention is a breakthrough over the current probe head and thus breaks the bottleneck limiting the process of fabricating the same, such that the method disclosed in the invention enables the probe head to be fabricated by a cheaper and less complicated process with more probes to be arranged in a unit area.

Method of fabricating vertical probe head

An electric generator is disclosed, which comprises: at least a magnet, each having more than two poles; and at least a claw-pole set, each being composed of an inner claw-pole and an outer claw-pole; wherein, the inner claw-pole and the outer claw-pole are interlaced arranged and used for guiding magnetic flux; the inner claw-pole is connected to an iron core whose outer diameter is smaller than the magnet and thus the loop of the inner claw-pole and the outer claw-pole is conducted; the core is winded by a solenoid coil; the number of claws of the inner claw-pole is the half of the pole number of the magnet while the outer claw-pole is the same, so that, as the magnet is move relative to the claw-pole set, the magnetic flux passing through the solenoid coil will change continuously and thus an induction electromotive force is generated.

Electric generator

This invention is a method for die separation of a wafer by ion implantation, wherein the die spacing is reduced and the die separation precision reaches a sub-micron level. The separated dies are obtained by wafer splitting after heat treatment and have cleavage roughness at a nano-meter level. The process method comprises the steps of: placing a wafer into an ion implanter; implanting gaseous ions into the split lines of said wafer with assistance of a mask; performing heat treatment of said wafer that is to split at the split lines; and completing die separation.

Method for die separation of a wafer by ion implantation

Growing community of inventors

Hsinchu, Taiwan

Fuh-Yu Chang