Differentiation of Yeast-Inoculated and Uninoculated Tomatoes Using Fluorescence Spectroscopy Combined with Machine Learning
, Vanya Slavova 2, Kadir Sabanci 3, Muhammet Fatih Aslan 3, Veselina Masheva 4 and Mariana Petkova 5
Round 1
Reviewer 1 Report
Dear Authors,
This paper discusses the differentiation of yeast-inoculated and uninoculated tomatoes by fluorescence spectroscopy and machine learning technology. This article is of high quality. However, this article needs a couple of improvements before acceptance. My comments are listed below:
Abstract: add conclusion after results.
Line 136-143, this section should provide the study purpose rather than the results.
Section 2.1. Material: Please provide a representative figure.
Figure 2 should be smaller to reduce space of journal.
Section 2.4. Please add details (model, company, country) in the entire paper, such as the AvaSpec- 198 ULS2048CL-EVO used.
The font of Table notes should be smaller.
Author Response
Comment 1. Abstract: add conclusion after results.
Response 1. Thanks for your remarkable opinion. The following statement has been added to the end of the abstract.
|
The results of the study show that the fluorescence spectroscopy data are strongly representative of tomato species. ML methods fed with these data provide high-performance discrimination. |
Comment 2. Line 136-143, this section should provide the study purpose rather than the results.
Response 2. Thanks for your comments. The purpose of the study is stated in the specified lines.
|
This study aims to classify the investigated tomatoes for different purposes using ML algorithms. Unlike previous studies, this study will present an ML-based application to discriminate between non-inoculated and yeast-inoculated tomato samples. The types of tomatoes used are ‘local dwarf’, ‘Ideal’ and ‘Picador’. They were examined spectrally using fluorescence spectroscopy and the obtained spectra were analyzed with six different ML algorithms such as Hoeffding Tree, PART, IBk, Filtered Classifier, Logistic and Bayes Net. The results of the analysis are expected to demonstrate the strong ability of the different ML methods to discriminate between non-inoculated and yeast-inoculated tomato cultivars. The contributions of this study can be summarized as follows.
· Various ML methods are used for the analysis of spectroscopic data.
· Computer aided systems are used to distinguish uninoculated and yeast-inoculated tomato samples.
· Different ML methods distinguish with high accuracy between uninoculated and yeast-inoculated tomato varieties.
|
Comment 3. Section 2.1. Material: Please provide a representative figure.
Response 3. Thank you for the good advice. Representative images of each species are included. The revision is as follows.
|
…The sample images representing the ‘local dwarf’, ‘Picador’ and ‘Ideal’ types used in this study are as in Figure 1.
|
|||||||||
Comment 4. Figure 2 should be smaller to reduce space of journal.
Response 4. We are grateful for your attention and interest. This figure has been redrawn, taking up less space as seen below. Please also see the revised article.
|
Figure 3. Schematic approach to fluorescence spectroscopic data acquisition and analysis. |
Comment 5. Section 2.4. Please add details (model, company, country) in the entire paper, such as the AvaSpec- 198 ULS2048CL-EVO used.
Response 5. We are grateful for your attention and interest. The revision is as follows. Please also see the revised article.
|
Also, the portable spectrometer model AvaSpec-ULS2048CL-EVO was used. AvaSpec-ULS2048CL-EVO is a portable spectrometer manufactured by Avantes - Apeldoorn(Netherlands) Using CMOS instead of conventional CCD technology, this spectrometer owes its key advantage over others with a similar configuration to the dominant position of the CMOS detector in its construction. technologies such as CMOS have evolved and become a suitable alternative. AvaSpec-ULS2048CL-EVO offers the latest technology providing high signal sensitivity. The spectrometer is combined with the latest electronics in the industry AS-7010. By purchasing AvaSpec-ULS0248CL-EVO, a universal device is purchased, including USB3.0 communication with 10 times higher speed compared to USB2 and a second communication port that offers Gigabit Ethernet for network integration and long-distance communication capability. In addition to high-speed communication options, the spectrometer also comes with a fast microprocessor and 50 times more memory capacity, which will help store a large batch of spectra.The tomato samples were placed on a duralumin stand reducing aberrations and allowing the emission of fluorescent signals of better quality (Figure 2). |
Comment 6. The font of Table notes should be smaller.
Response 6. We are grateful for your attention and interest. The font of table notes has been made smaller. Please see the revised article.
Author Response File: 
Author Response.docx
Reviewer 2 Report
This paper is based on Differentiation of yeast-inoculated and uninoculated tomatoes using fluorescence spectroscopy combined with machine learning. Thus, this paper is directly related to the theme of this journal.
Overall, the paper is organized properly; the concept and future research directions are extensively explained.
1. Paper contains few grammar mistakes which will be cooperated in final version.
2. Contribution of paper must be given in bullets
3. Add organization or reminder of paper in the end of introduction for reader guidance
4. It’s better to add results of table 3 in graphs that will be easy for readers
5. Only 33 references are added in paper, so to attract readers add few latest references related to, which is mentioned below
Hui He, Muhammad Shafiq, and Asiya Khan. "Assessment of quality of experience (QoE) of image compression in social cloud computing." Multiagent and Grid Systems 14, no. 2 (2018): 125-143.
Karim, Shahid, Ye Zhang, and Muhammad Rizwan Asif. "Image processing based proposed drone for detecting and controlling street crimes." In 2017 IEEE 17th International Conference on Communication Technology (ICCT), pp. 1725-1730. IEEE, 2017.
Karim, Shahid, Ye Zhang, Shoulin Yinand Ali Anwar Brohi. "Impact of compressed and down-scaled training images on vehicle detection in remote sensing imagery." Multimedia Tools and Applications 78, no. 22 (2019): 32565-32583.
Author Response
RESPONSES TO REVIEWER 2
Comment 1. Paper contains few grammar mistakes which will be cooperated in final version.
Response 1. Thanks for your remarkable comment. Grammar mistakes have been corrected. Please see the revised article.
Comment 2. Contribution of paper must be given in bullets.
Response 2. Thanks for your remarkable comment. The contributions of the study are included in the revised article as follows.
|
The contributions of this study can be summarized as follows.
· Various ML methods are used for the analysis of spectroscopic data.
· Computer aided systems are used to distinguish uninoculated and yeast-inoculated tomato samples.
· Different ML methods distinguish with high accuracy between uninoculated and yeast-inoculated tomato varieties.
|
Comment 3. Add organization or reminder of paper in the end of introduction for reader guidance.
Response 3. Thanks for your remarkable comment. A paragraph for paper organization has been added at the end of the introduction. This revision is shown below.
|
This paper is organized as follows. After the introduction, section 2 explains in detail the tomato species used, growing conditions, information on the inoculation of tomatoes, fluorescence spectroscopy data obtained from tomatoes, and discriminant analysis. Section 3 interprets the results obtained from the different ML methods as a result of the classification of each tomato species. Finally, section 4 evaluates and concludes the study. |
Comment 4. It’s better to add results of table 3 in graphs that will be easy for readers
Response 4. Thanks for your remarkable comment. Table 3 entitled “The discrimination performance metrics of control and yeast-inoculated tomato ‘Ideal’ (COE0160).” presents the results of the classification of yeast-inoculated and uninoculated samples for one (‘Ideal’) of three tested tomato types. Table 1 shows the results for local dwarf tomatoes and Table 2 for ‘Pikador’. All Tables 1, 2, and 3 include the same classification metrics such as confusion matrices, average accuracy, True Positive Rate (TP Rate), Precision, Receiver Operating Characteristic Area (ROC Area), Precision-Recall Area (PRC Area), F-Measure, and Matthews Correlation Coefficient (MCC). Therefore, to compare the obtained results for all three types of tomatoes, if the results for local dwarf and ‘Pikador’ are in tables (1 and 2), the results for the ‘Ideal’ variety are also presented in table (3).
Comment 5. Only 33 references are added in paper, so to attract readers add few latest references related to, which is mentioned below.
- a) Hui He, Muhammad Shafiq, and Asiya Khan. "Assessment of quality of experience (QoE) of image compression in social cloud computing." Multiagent and Grid Systems 14, no. 2 (2018): 125-143.
- b) Karim, Shahid, Ye Zhang, and Muhammad Rizwan Asif. "Image processing based proposed drone for detecting and controlling street crimes." In 2017 IEEE 17th International Conference on Communication Technology (ICCT), pp. 1725-1730. IEEE, 2017.
- c) Karim, Shahid, Ye Zhang, Shoulin Yinand Ali Anwar Brohi. "Impact of compressed and down-scaled training images on vehicle detection in remote sensing imagery." Multimedia Tools and Applications 78, no. 22 (2019): 32565-32583.
Response 5. Thanks for your remarkable comment. New citations including the above works have been added to the revised article. Please see the revised article.
Author Response File: Author Response.docx
Reviewer 3 Report
This paper is promising. However, the authors should compare with other results about this topic using other methods. This paper has not provided any funding information. Is it correct? These two kinds of tomatoes were classified using this method. However, the authors should compare with other methods about the detection cost. If the cost is high, it is not promising ever if this method is correct. And other sides of this method are also compared, such as complex.
Author Response
RESPONSES TO REVIEWER 3
Comment 1. This paper is promising. However, the authors should compare with other results about this topic using other methods.
Response 1. Thanks for your remarkable comment. The application of fluorescence spectroscopy has been justified as follows. However, in future research, the authors plan to use image processing as another non-destructive quality assessment technique.
|
Fluorescence spectroscopy is widely used in the food industry for quantitative analysis. It is sensitive and specific enough to detect even small concentrations of the compounds [25,26]. It can, for example, detect changes in the structures of proteins, carbohydrates and lipids in oils. This is useful for verifying the authenticity of food products [27]. Advances in fiber optic technology offer exceptional opportunities to develop a wide range of highly sensitive fibre optic sensors in many new areas of ap-plications. The fiber-optic components are successfully adapted to compilations with elements of micro-optics such as lenses, mirrors, prisms, gratings, etc. [28,29]. Fluorescence spectroscopy in the agricultural sciences is applied to the analysis of tomatoes [30] and cereals [31]. Their qualification by means of this technique is performed by grouping objects with similar characteristics for establishing methods related to their classification. |
Comment 2. This paper has not provided any funding information. Is it correct?
Response 2. Thanks for your remarkable comment. That's right, there is no source of finance.
Comment 3. These two kinds of tomatoes were classified using this method. However, the authors should compare with other methods about the detection cost. If the cost is high, it is not promising ever if this method is correct. And other sides of this method are also compared, such as complex.
Response 3. Thanks for your remarkable comment. The developed procedures to distinguish tomato samples were non-destructive, objective, easy, and fast. The spectroscopic techniques require no sample preparation or using chemical reagents and can be used to directly estimate the sensory quality. It has been included in the manuscript as follows:
|
This study presents an application of Machine Learning (ML) methods using fluorescent spectroscopic data to distinguish between uninoculated and yeast-inoculated tomato samples. Since traditional classification techniques are both cumbersome and difficult, computer vision and artificial intelligence methods, which offer easy, more accurate and fast solutions in classification studies of agricultural products, are frequently preferred recently. Among the artificial intelligence methods, ML includes a large number of algorithms with a robust discrimination ability if strong features are extracted from agricultural products. ML is a method that is aimed to improve itself by using experience and data [36]. Although it occupies a common area with artificial intelligence, it has also a standalone subject. ML algorithms have the purpose of predicting future data using previous data on a subject for which they are not specifically programmed. The previous data mentioned here are called training data in the literature. The accuracy of the model created using training data is examined using test data [37]. |
Author Response File: Author Response.docx
