Ultrasound based glass fragments detection in glass containers filled with beverages using neural networks and short time Fourier transform

Zhao, Bosen; Yang, Pei; Basir, Otman; Mittal, G.S.

doi:10.1016/j.foodres.2006.01.008

Cited by 8 publications

(3 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The pressure ratio was the lowest value when there were no foreign bodies inside the container for frequencies up to 8 MHz. They also attempted to detect a glass fragment using an ultrasonic sensor based on a combination of the radial basis function neural network algorithm and the short‐time Fourier transform (Zhao, Yang, Basir, & Mittal, ). The input features consisted of parameters related to the glass fragment such as size, position, orientation and the incident angle of the ultrasonic beam.…”

Section: Noninvasive Techniquesmentioning

confidence: 99%

Noninvasive techniques for detection of foreign bodies in food: A review

Khairi

Ibrahim

Yunus

et al. 2018

J Food Process Engineering

View full text Add to dashboard Cite

Foreign body in food is a matter of concern in food industry as it determines the safety and quality of the product. This issue leads to anxiety as it is hazardous if it is accidentally being consumed. Therefore, foreign body detection is urgently needed. Noninvasive techniques are attractive as they can be used to evaluate without altering the originality of the food products in terms of ingredient and structure. With the growing interest in this subject, this article reviews the development of noninvasive techniques including X‐ray, thermal imaging, near‐infrared spectroscopy, hyperspectral imaging, ultrasonic, and terahertz. The principle and application of each technique are elaborated. The performances and limitations from previous studies in several types of food applications are analyzed. In addition, future trends and challenges encountered with these techniques are highlighted. It is envisaged that the information gathered in this article will be a valuable source of information for researchers working in this topic. Practical applications The existence of foreign bodies in food causes degradation of their quality and safety. The implementation of the noninvasive technique as a monitoring tool have received an encouraging response from the manufacturers which endeavors to maintain their credibility and reputation. This review serves to provide an up‐to‐date development on the noninvasive technique for detecting foreign bodies which is useful for academicians and researchers especially from the food processing industries.

show abstract

Section: Noninvasive Techniquesmentioning

confidence: 99%

Noninvasive techniques for detection of foreign bodies in food: A review

Khairi

Ibrahim

Yunus

et al. 2018

J Food Process Engineering

View full text Add to dashboard Cite

show abstract

“…However, the pressure ratio values for all conditions became close to each other at frequencies of 9 MHz and 10 MHz. Zhao et al [80] investigated the detection of a glass fragment using an ultrasonic sensor based on a combination of the radial basis function neural network algorithm and short-time Fourier transform. Twenty-two variables were used as inputs, which are the features representing the glass fragment conditions such as size, position, orientation and the incident angle of the ultrasonic beam.…”

Section: Food Safetymentioning

confidence: 99%

Contact and non-contact ultrasonic measurement in the food industry: a review

Khairi

Ibrahim

Yunus

et al. 2015

Meas. Sci. Technol.

View full text Add to dashboard Cite

The monitoring of the food manufacturing process is vital since it determines the safety and quality level of foods which directly affect the consumers’ health. Companies which produce high quality products will gain trust from consumers. This factor helps the companies to make profits. The use of efficient and appropriate sensors for the monitoring process can also reduce cost. The food assessing process based on an ultrasonic sensor has attracted the attention of the food industry due to its excellent capabilities in several applications. The utilization of low or high frequencies for the ultrasonic transducer has provided an enormous benefit for analysing, modifying and guaranteeing the quality of food. The contact and non-contact ultrasonic modes for measurement also contributed significantly to the food processing. This paper presents a review of the application of the contact and non-contact mode of ultrasonic measurement focusing on safety and quality control areas. The results from previous researches are shown and elaborated.

show abstract

“…Time-frequency diagrams calculated from STFT of various types of radar signals were used with a CNN to discriminate radar signals in another study [12,13]. The time-frequency diagram of ultrasonic signals of broken glass in glass containers can be used to identify glass fragments by CNN [14]. Lee et al [15] distinguished human emotions by time-frequency diagrams of pupil size change and movement data by the CNN.…”

Section: Introductionmentioning

confidence: 99%

A Novel Classification Method for Flutter Signals Based on the CNN and STFT

Duan

Zheng

Liu

2019

International Journal of Aerospace Engineering

View full text Add to dashboard Cite

Necessary model calculation simplifications, uncertainty in actual wind tunnel test, and data acquisition system error altogether lead to error between a set of actual experimental results and a set of theoretical design results; wind tunnel test flutter data can be utilized to feedback this error. In this study, a signal processing method was established to use the structural response signals from an aeroelastic model to classify flutter signals via deep learning algorithm. This novel flutter signal processing and classification method works by combining a convolutional neural network (CNN) with time-frequency analysis. Flutter characteristics are revealed in both time and frequency domains, which are harmonic or divergent in the time series; the flutter model energy is singular and significantly increases in the frequency view, so the features of the time-frequency diagram can be extracted from the dataset-trained CNN model. As the foundation of the subsequent deep learning algorithm, the datasets are placed into a collection of time-frequency diagrams calculated by short-time Fourier transform (STFT) and labeled with two artificial states, flutter or no flutter, depending on the source of the signal measured from a wind tunnel test on the aeroelastic model. After preprocessing, a cross-validation schedule is implemented to update (and optimize) CNN parameters though the trained dataset. The trained models were compared against test datasets to validate their reliability and robustness. Our results indicate that the accuracy rate of test datasets reaches 90%. The trained models can effectively and automatically distinguish whether or not there is flutter in the measured signals.

show abstract

Ultrasound based glass fragments detection in glass containers filled with beverages using neural networks and short time Fourier transform

Cited by 8 publications

References 7 publications

Noninvasive techniques for detection of foreign bodies in food: A review

Noninvasive techniques for detection of foreign bodies in food: A review

Contact and non-contact ultrasonic measurement in the food industry: a review

A Novel Classification Method for Flutter Signals Based on the CNN and STFT

Contact Info

Product

Resources

About