Oscar Chang

Dr. Oscar Chang


Research Focus & Metrics

Academic Career Length

46Years (1981-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
29 August 2016 in IEEE Latin America Transactions
Verified
Journal Article
Closed access
Patricia ConstantePatricia Constante
Patricia Constante
AGA. Gordon
A. Gordon
OCOscar Chang
Oscar Chang
EPE. Pruna
E. Pruna
FAFausto Acuna
Fausto Acuna
IEI. Escobar
I. Escobar
This research presents novel artificial vision techniques applied to the detection of features for strawberries used in the food industry. For this purpose, a computer vision system based in artificial neural networks is used, organized as a deep architecture and trained with noise compensated learning. This combination originates a strong network - object relations which makes possible the recognition of complex strawberry features under changing conditions of lightning, size and orientation. The programming uses OpenCV libraries and fruits databases captured with a webcam. The images used to train the Artificial Neural Network are defined with canny edge detection and a moving region of interest (ROI). After training, the network recognizes important features such as shape, color and anomalies. The system has been tested in real time with real images.

23 February 2017 in Journal of Artificial Intelligence and Soft Computing Research
Verified
Journal Article
Open access
PDF file
OCOscar Chang
Oscar Chang
Patricia ConstantePatricia Constante
Patricia Constante
AGAndrés Gordon
Andrés Gordon
MSMarco Singaña
Marco Singaña
This work proposes a deep neural net (DNN) that accomplishes the reliable visual recognition of a chosen object captured with a webcam and moving in a 3D space. Autoencoding and substitutional reality are used to train a shallow net until it achieves zero tracking error in a discrete ambient. This trained individual is set to work in a real world closed loop system where images coming from a webcam produce displacement information for a moving region of interest (ROI) inside the own image. This loop gives rise to an emergent tracking behavior which creates a self-maintain flow of compressed space-time data. Next, short term memory elements are set to play a key role by creating new representations in terms of a space-time matrix. The obtained representations are delivery as input to a second shallow network which acts as “recognizer”. A noise balanced learning method is used to fast train the recognizer with real-world images, giving rise to a simple and yet powerful robotic eye, with a slender neural processor that vigorously tracks and recognizes the chosen object. The system has been tested with real images in real time.

1 January 2023 in Uniciencia
Verified
Journal Article
Open access
PDF file
LRLeo Thomas Ramos
Leo Thomas Ramos
OCOscar Chang
Oscar Chang
[Objective] The moment Russia officially invaded Ukraine, the world experienced a period of tension and uncertainty. As a social release valve digital communication, channels increased their number of users and activity, generating a large amount of data. Twitter, in particular, being one of the most popular channels for sharing information and opinions, exploded with activities related to this historical moment. And as with many other social events, such as COVID-19, this social network became one of the main sources of information, opinion, and knowledge. This paper analyzes sentiments in tweets related to the armed conflict between Russia and Ukraine. [Methodology] The analyzed dataset contains tweets from January 1, 2022, through March 3, 2022, and was collected using event-related hashtags. In total, 603,552 tweets in English and 1,664 in Russian were analyzed. To perform emotion classification, DistilRoBERTa variant and the pre-trained XLM-RoBERTa-Base model were used, respectively. English tweets were classified into seven emotions: anger, disgust, fear, joy, neutral, sadness, and surprise. Russian tweets were classified into positive negative, and neutral polarities. [Results] The results showed that most English tweets convey fear and anger as predominant feelings, reaching 32.08% and 15.18% of the total tweets analyzed, respectively. Regarding tweets in Russian, the majority presented negative polarity, with 86.83% of the total. Some of the most recurrent phrases in the analysis allude to support for Ukraine and call for a halt to the war. Likewise, phrases of concern for the crisis, weapons, and fatalities are recurrent. [Conclusion] As expected, most people were concerned about the armed conflict and upset and angry about its consequences. Future works could use more tweets to improve the analysis and increase the time range to be studied. The analysis could also be segmented to study the sentiments of tweets according to different groupings and compare them with other societies, for instance, tweets could be segmented by country and analyzed accordingly.

31 October 2020
Verified
Book Chapter
Closed access
PDF file
RVRonny Velastegui
Ronny Velastegui
LVLuis Fernando Zhinin Vera
Luis Fernando Zhinin Vera
GPGissela E. Pilliza
Gissela E. Pilliza
OCOscar Chang
Oscar Chang

19 November 2002
Verified
Conference Paper
Closed access
DMD. Marcano
D. Marcano
FDF. Duran
F. Duran
OCOscar Chang
Oscar Chang

31 October 2020
Verified
Book Chapter
Closed access
PDF file
FQFrancisco Quinga-Socasi
Francisco Quinga-Socasi
LVLuis Fernando Zhinin Vera
Luis Fernando Zhinin Vera
OCOscar Chang
Oscar Chang