Non-Destructive Monitoring of Sweet Pepper Samples After Selected Periods of Lacto-Fermentation

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

Agriculture

Non-destructive monitoring of sweet pepper samples after selected periods of lacto-fermentation

The work reports the application of image analysis to classify yellow and red pepper according to fermentation stage. The text is well written, the methods used are clearly described, and the features to assess model performance are appropriate. There is a lack of background to justify the work, and the lack of validation dataset limits the work. More detailed information are provided below:

 

Introduction

There is a comprehensive background about the importance of bell pepper regarding its nutritional attributes, but there is no mention about the need to use image analysis, or any other aspect of computer vision, AI, etc. Paragraphs 2, 3 and 5 in the introduction could be removed, and an introduction about the need to use AI in pepper classification should be provided.

In addition, the authors could provide information about the need to ferment pepper, and also the most recent advances in AI models applied to image in agricultural and food products, to justify the current work. .

Please note that authors can evaluate the references suggested based on their relevance and appropriateness in supporting the current work. The authors have the freedom to make independent judgments and ensure the integrity and clarity of the work, as they are not obligated to incorporate any citation which is not deemed relevant or appropriate for the current research. The final opinion of this reviewer will not be contingent upon adding those references.

Material and methods

Please present the demand for red and yellow pepper, why are they important? What is the need to identify its fermentation step?

Please clarify the number of samples analyzed and the number of images used in the models.

Please present the number of images used in model calibration and external prediction. There should be external prediction for a different set of samples.

Regarding the models used, please justify its choice.

Results

Considering the models used, please justify the application of models without information about the features responsible for its success. The references provided may help the authors to compare the current results with explainable methods, as it could help understand what is being identified by the model.

Lines 285-307: Try to use the reported studies to address the success/failure of the current work, instead of only citing previous results. This could help enrich the discussion providing insights for other researchers about the methods used and its usefulness.

Comments on the Quality of English Language

Appropriate, minor changes required

Author Response

Response to Reviewer 1 Comments

                           

Thank you very much for taking the time to review this manuscript. Please find the detailed responses below and the corresponding revisions/corrections highlighted in red in the re-submitted files.

 

Point-by-point response to Comments and Suggestions for Authors

Comment 1: Agriculture

Non-destructive monitoring of sweet pepper samples after selected periods of lacto-fermentation

The work reports the application of image analysis to classify yellow and red pepper according to fermentation stage. The text is well written, the methods used are clearly described, and the features to assess model performance are appropriate. There is a lack of background to justify the work, and the lack of validation dataset limits the work. More detailed information are provided below:

Response 1: Thank you very much for reading the paper carefully and your valuable comments. Your suggestions have been considered and the manuscript has been improved.

The background to justify the work has been included in the Introduction. This section has been significantly revised and supplemented with information regarding the demand for fermented foods and the benefits of fermentation, as well as the need to monitor the changes in the properties of fermented products using non-destructive and objective approaches using computer vision and machine learning. The mentioned literature data considered as the background for the current study provided a perspective for carrying out research towards the non-destructive evaluation of the quality of food products using image analysis and machine learning and justified the current work. The background to justify the work has been included in lines 49-91 and marked in red.

As mentioned in the revised version of the manuscript, the work included the validation datasets. Validation was performed using separate datasets from those used for training. It provided objective results. It has been explained in the manuscript as follows:

“The test mode of 10-fold cross-validation was applied. It means that separate datasets were used for validation than those for training. Each dataset included 800 cases (100 cases for each of the 8 pepper classes). The datasets were randomly divided into 10 parts. Nine parts were considered as the training sets and one part was the validation set. The process was repeated 10 times with different validation sets and the overall results were the averages of 10 estimations. Ten folds ensured high results of classification metrics and reproducibility of results.” [lines 166-172]

 

Introduction

Comment 2: There is a comprehensive background about the importance of bell pepper regarding its nutritional attributes, but there is no mention about the need to use image analysis, or any other aspect of computer vision, AI, etc. Paragraphs 2, 3 and 5 in the introduction could be removed, and an introduction about the need to use AI in pepper classification should be provided.

Response 2: The authors agree with this valuable comment. The Introduction has been corrected and improved. Paragraphs 2, 3, and 5 have been removed. Information about the use of image analysis, computer vision, and artificial intelligence for the classification of fermented food products and the need to use this approach to classify fermented pepper has been added as follows:

“The quality of processed agricultural food can be monitored using computer vision systems. These systems are important for the accurate and consistent inspection of external quality parameters [21]. For example, the effect of lacto-fermentation for 1 week and 3 months was assessed on zucchini flesh obtaining overall accuracies of up to 99.33% (IBk). Additionally, it was found that the fresh sample was completely different in terms of image texture from lacto-fermented samples [22]. For the carrot, models built based on image texture parameters provided 100% accuracy in distinguishing fresh carrot slices and samples lacto-fermented for 6 months in the case of the Multilayer Perceptron machine learning algorithm [23]. Moreover, fermentation level using artificial intelligence and computer vision was investigated for cocoa beans by Anggraini et al. [24], who found an accuracy of 94% (Multilayer Perceptron) for distinguishing fermented and unfermented cocoa and concluded that the applied approach could predict cocoa bean fermentation rate. Oliveira et al. [25] distinguished fully fermented, partially fermented, under fermented, and unfermented cocoa beans (cut test) using computer vision obtaining 0.93 accuracy. Machine learning algorithms based on image features were used by Bhargava et al. [26] for the grading of tea samples, such as fermented, over-fermented, and under-fermented. The algorithms detected the tea fermentation level with an accuracy reaching 98.75% (SVM), 89.72% (SRC), and 87.39% (k-NN). In the case of black tea, machine learning algorithms were also applied in quality assurance and detecting the degree of fermentation based on electrical properties by Zhu et al. [27] resulting in an accuracy of up to 100% (Random Forest). The reported results indicated the usefulness of image features and artificial intelligence for the classification of fermented products with very high accuracy reaching 100%. The above literature data provided a perspective for carrying out research towards the non-destructive evaluation of the quality of food products using image analysis and machine learning and justified the current work. It was assumed that this approach could be used for non-destructive and objective monitoring of pepper samples during fermentation.

In this context, this study aimed at non-destructive distinguishing bell pepper samples after specific periods of lacto-fermentation to assess the changes occurring during the process.” [lines 65-94]

 

Comment 3: In addition, the authors could provide information about the need to ferment pepper, and also the most recent advances in AI models applied to image in agricultural and food products, to justify the current work. .

Response 3: The manuscript has been corrected according to this important comment. Information about the need to ferment pepper has been provided as follows:

“In recent years, there has been a steady increase in demand for fermented plant products, which have been recognized as functional products that benefit human health in many ways [16]. More than 2,300 pepper cultivars are registered on the European market, varying in size, color, shape, spiciness, juiciness, or flesh thickness. Commercial cultivars of pepper are available with different skin colors - red, green, orange, yellow, or purple. The differences in fruit color are due to the ability to synthesize chlorophyll or carotenoids and the degree of maturity of the fruit [17].

 According to Blanco-Ríos [18], the nutritional value of peppers depends, among other factors, on the color of the fruit (cultivar), growing conditions, and post-harvest treatment. The active substances found in pepper fruits have antioxidant (high content of vitamins mainly A, C, and E, as well as minerals and polyphenols), anti-inflammatory, anti-diabetic, antimicrobial, and immunomodulation effects. Bell pepper fruit consists of 80-90% water and therefore has a relatively short shelf life compared, for example, to root vegetables. Shortly after harvesting, the fruit wilts and quickly loses its commercial value [19]. The processing of pepper fruit, especially the use of a lactic fermentation process with lactic acid bacteria, contributes to a change in taste, texture, and aroma [20]. [lines 49-64]

 

Information about the most recent advances in AI models applied to images in agricultural and food products to justify the current work has been provided as follows:

“The quality of processed agricultural food can be monitored using computer vision systems. These systems are important for the accurate and consistent inspection of external quality parameters [21]. For example, the effect of lacto-fermentation for 1 week and 3 months was assessed on zucchini flesh obtaining overall accuracies of up to 99.33% (IBk). Additionally, it was found that the fresh sample was completely different in terms of image texture from lacto-fermented samples [22]. For the carrot, models built based on image texture parameters provided 100% accuracy in distinguishing fresh carrot slices and samples lacto-fermented for 6 months in the case of the Multilayer Perceptron machine learning algorithm [23]. Moreover, fermentation level using artificial intelligence and computer vision was investigated for cocoa beans by Anggraini et al. [24], who found an accuracy of 94% (Multilayer Perceptron) for distinguishing fermented and unfermented cocoa and concluded that the applied approach could predict cocoa bean fermentation rate. Oliveira et al. [25] distinguished fully fermented, partially fermented, under fermented, and unfermented cocoa beans (cut test) using computer vision obtaining 0.93 accuracy. Machine learning algorithms based on image features were used by Bhargava et al. [26] for the grading of tea samples, such as fermented, over-fermented, and under-fermented. The algorithms detected the tea fermentation level with an accuracy reaching 98.75% (SVM), 89.72% (SRC), and 87.39% (k-NN). In the case of black tea, machine learning algorithms were also applied in quality assurance and detecting the degree of fermentation based on electrical properties by Zhu et al. [27] resulting in an accuracy of up to 100% (Random Forest). The reported results indicated the usefulness of image features and artificial intelligence for the classification of fermented products with very high accuracy reaching 100%. The above literature data provided a perspective for carrying out research towards the non-destructive evaluation of the quality of food products using image analysis and machine learning and justified the current work. It was assumed that this approach could be used for non-destructive and objective monitoring of pepper samples during fermentation. [lines 65-91]

 

Comment 4: Please note that authors can evaluate the references suggested based on their relevance and appropriateness in supporting the current work. The authors have the freedom to make independent judgments and ensure the integrity and clarity of the work, as they are not obligated to incorporate any citation which is not deemed relevant or appropriate for the current research. The final opinion of this reviewer will not be contingent upon adding those references.

Response 4: The authors are grateful for this comment. Relevant references have been added to the Introduction as follows:

21. Zhang, B.; Guo, N.; Huang, J.; Gu, B.; Zhou, J. Computer Vision Estimation of the Volume and Weight of Apples by Using 3D Reconstruction and Noncontact Measuring Methods. J. Sens. 2020, 2020, 5053407.
22. Ropelewska, E.; Sabanci, K.; Aslan, M.F. The effect of lacto-fermentation over time on the changes in zucchini flesh quality assessed using machine learning models based on image textures. Journal of Food Process Engineering 2023, 46(12), e14496.
23. Ropelewska, E. Distinguishing lacto-fermented and fresh carrot slice images using the Multilayer Perceptron neural network and other machine learning algorithms from the groups of Functions, Meta, Trees, Lazy, Bayes and Rules. Eur Food Res Technol 2022, 248, 2421–2429.
24. Anggraini, C.D.; Putranto, A.W.; Iqbal, Z.; Firmanto, H.; Riza, D.F.A. Preliminary study on development of cocoa beans fermentation level measurement based on computer vision and artificial intelligence. In, 2021 International Conference on Green Agroindustry and Bioeconomy; 924 of IOP Conference Series, Earth and Environmental Science. IOP Publishing. 2021, 1–8.
25. Oliveira, M.M.; Cerqueira, B.V.; Barbon, Jr S.; Barbin, D.F., Classification of fermented cocoa beans (cut test) using computer vision. J. Food Compos. Anal. 2021, 97, 103771.
26. Bhargava, A.; Bansal, A.; Goyal, V. et al. Machine learning & computer vision-based optimum black tea fermentation detection. Multimed Tools Appl 2023, 82, 43335–43347.
27. Zhu, H.K.; Liu, F.; Ye, Y.; Chen, L.; Liu, J.Y.; Gui, A.H.; et al. Application of machine learning algorithms in quality assurance of fermentation process of black tea-based on electrical properties. Journal of Food Engineering 2019, 263, 165–172.

 

The reported information has been evaluated and the justification for conducting the current research was explained in the context of literature data as follows:

“The reported results indicated the usefulness of image features and artificial intelligence for the classification of fermented products with very high accuracy reaching 100%. The above literature data provided a perspective for carrying out research towards the non-destructive evaluation of the quality of food products using image analysis and machine learning and justified the current work. It was assumed that this approach could be used for non-destructive and objective monitoring of pepper samples during fermentation.

In this context, this study aimed at non-destructive distinguishing bell pepper samples after specific periods of lacto-fermentation to assess the changes occurring during the process.” [lines 85-94]

 

Material and methods

Comment 5: Please present the demand for red and yellow pepper, why are they important? What is the need to identify its fermentation step?

Response 5: The importance of pepper with different colors of fruit and the demand for fermented pepper have been explained as follows:

“In recent years, there has been a steady increase in demand for fermented plant products, which have been recognized as functional products that benefit human health in many ways [16]. More than 2,300 pepper cultivars are registered on the European market, varying in size, color, shape, spiciness, juiciness, or flesh thickness. Commercial cultivars of pepper are available with different skin colors - red, green, orange, yellow, or purple. The differences in fruit color are due to the ability to synthesize chlorophyll or carotenoids and the degree of maturity of the fruit [17].” [lines 49-55]

 

The need to identify its fermentation step has been explained as follows:

“In this context, this study aimed at non-destructive distinguishing bell pepper samples after specific periods of lacto-fermentation to assess the changes occurring during the process. Peppers with different fruit colors can have different properties and, consequently, processing, including fermentation, may take different courses. Therefore, pepper raw materials with completely different colors were selected. The identification of the pepper fermentation step is important and can be needed to determine after what period the greatest changes in the structure of the flesh of fermented samples occur and when the changes are the smallest. Based on this information the conditions and duration of fermentation can be established to obtain the most desired product.” [lines 92-100]

 

Comment 6: Please clarify the number of samples analyzed and the number of images used in the models.

Response 6: It has been clarified as follows:

“The image acquisition was performed in one hundred repetitions. For each pepper class, ten images with ten pieces in each image were obtained. During the image processing, each image was cropped into ten images including one pepper piece in each. Therefore in total one hundred images were obtained for each group.” [lines 139-143]

 

Comment 7: Please present the number of images used in model calibration and external prediction. There should be external prediction for a different set of samples.

Response 7: The prediction process has been presented as follows:

“The test mode of 10-fold cross-validation was applied. It means that separate datasets were used for validation than those for training. Each dataset included 800 cases (100 cases for each of the 8 pepper classes). The datasets were randomly divided into 10 parts. Nine parts were considered as the training sets and one part was the validation set. The process was repeated 10 times with different validation sets and the overall results were the averages of 10 estimations. Ten folds ensured high results of classification metrics and reproducibility of results.” [lines 166-172]

 

Comment 8: Regarding the models used, please justify its choice.

Response 8: It has been justified as follows:

“The choice of the models used for the classification of red bell pepper samples before and after various periods of lacto-fermentation was justified by providing the highest accuracies and values of the other above-mentioned metrics. The most successful for distinguishing red bell pepper samples were models built using IBk, WiSARD, and Random Committee algorithms. Whereas in the case of the classification of yellow bell pepper samples, LMT and IBk algorithms provided the highest results.” [lines 190-195]

 

Results

Comment 9: Considering the models used, please justify the application of models without information about the features responsible for its success. The references provided may help the authors to compare the current results with explainable methods, as it could help understand what is being identified by the model.

Response 9: The features with the highest discriminative power responsible for the success of models have been presented in the manuscript as follows:

“The following image texture parameters of red bell pepper were characterized by the highest discriminative power: RSGArea, RS5SH1SumVarnc, RS5SH1DifEntrp, RS5SH3SumVarnc, RS5SZ3SumVarnc, RATeta1, GHPerc01, BS4RNGLevNonU, LHPerc01, XHPerc01, XSGNonZeros, XSGPerc01, XS5SH1SumVarnc, XS5SZ1AngScMom, UHPerc10, and UHDomn01.” [lines 201-206]

 

“In the case of the yellow bell pepper, the image textures with the highest discriminative power, which were the most important for building the most successful models were: RHMaxm01, RS5SV1SumVarnc, RS4RVLngREmph, GS5SZ5DifVarnc, LHSkewness, LHKurtosis, aHSkewness, aHKurtosis, aHDomn10, XHSkewness, YHMean, ZS5SZ3DifEntrp, US4RVGLevNonU, UASigma, SHPerc99, SS5SH5InvDfMom, SS5SN5SumVarnc, and SASigma.” [lines 281-286]

 

Comment 10: Lines 285-307: Try to use the reported studies to address the success/failure of the current work, instead of only citing previous results. This could help enrich the discussion providing insights for other researchers about the methods used and its usefulness.

Response 10: The discussion has been completely rewritten and revised to include this important comment. 

A new discussion of the obtained results has been added as follows:

“In our study, the application of image analysis of pepper samples and machine learning models developed based on texture image parameters allowed for the non-destructive and objective determination of differences in samples before and after selected periods of lacto-fermentation. The image texture features were used to distinguish the samples. The image texture is a function of the spatial variation of the pixel brightness intensity. Image textures can carry important information about the structure of the evaluated samples. The quantitative analyses of texture parameters from pictures provide valuable insights into product quality [28, 38]. Therefore the changes in the quality of peppers during the fermentation were evaluated based on pictures in terms of flesh structure.

The determined correctness of classification for individual samples revealed the changes in the structure of the peppers during the process. It was observed that the classification accuracy depended on the pepper cultivar. In the case of red bell pepper, the average accuracy of distinguishing samples before and after 3, 7, 10, 14, 21, 28, and 56 days of lacto-fermentation reached 93%. Whereas the samples of yellow bell pepper were correctly classified with an average accuracy of up to 90%. It meant that the differences between samples were slightly greater for red pepper. Also, Janiszewska-Turak et al. [17] reported that the properties of fermented pepper depend on the cultivar.

Furthermore, the accuracy of distinguishing pepper samples depended on the machine learning algorithm used to build the models. Among the many tested algorithms from the groups of Functions, Bayes, Meta, Lazy, Trees, and Rules, the highest results were obtained for IBk (Lazy) – 93%, WiSARD (Functions) – 92%, and Random Committee (Meta) – 88% in the case of red bell pepper and LMT (Trees) – 90% and IBk (Lazy) – 89% for yellow bell pepper samples. Also, literature data reported the usefulness of image analysis and machine learning for the quality assessment of fermented pepper. In the previous study regarding red bell pepper, the use of IBk, SMO, Random Forest, Naive Bayes, Filtered Classifier, and JRip machine learning algorithms allowed for the distinguishing fresh and lacto-fermented samples for 6 months with an average accuracy of 99% [39].

The obtained results can be of practical application. A vision system that takes into account the use of imaging devices can be used to determine the correct course of the fermentation process. Additionally, knowing the course of fermentation for different cultivars, texture parameters of the flesh of pepper raw materials can be useful for planning the process. However, the applied approach also has limitations. The practical problem of fermentation is the fact that the course of the process and the quality of fermented products depend on the cultivar, degree of maturity, and physicochemical properties of the raw material [40,41]. To scale up the laboratory experiments to practical applications, it is necessary to collect more data for a larger number of cultivars, degrees of maturity, and growing seasons.

Although spontaneous fermentation is a reliable way of preserving fruit and vegetables, there is a risk of fermentation failure, undesirable and unpredictable changes in the quality properties, or insufficient inhibition of microorganisms causing product spoilage [42]. The application of spontaneous/natural fermentation has also disadvantages related to the fact that spoilage organisms can be easily introduced, fermentation strains are unknown, starting fermentation is difficult, and the results of the process are not controllable [43]. Therefore, in our study, slight differences could have occurred between individual technological repetitions, influencing the properties of the products and the accuracies of their classification. To avoid the problem of uncontrolled fermentation and ensure greater safety and higher quality of fermented products, selected starter cultures can be used to perform the process [44].” [lines 332-380]

 

Response to Comments on the Quality of English Language

Comment 1: Appropriate, minor changes required

Response 1: The authors are grateful for this comment. The manuscript has been checked thoroughly and English Language has been improved.

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

In this work, combined with machine learning models, image analysis was used to recognize differences in pepper samples after selected periods of lacto-fermentation. I think the manuscript is in the scope of the Agriculture and can be considered for the publication after minor revision. The following comments should be clarified before publish.

Comment 1: Can the quality be evaluated solely based on pictures of peppers during the fermentation period?

Comment 2: The author used multiple models to classify and recognize peppers, but the comparison of recognition results among different models should be discussed.

Comment 3: The applicability and scalability of the models should be discussed.

 

Comments on the Quality of English Language

 Minor editing of English language required.

Author Response

Response to Reviewer 2 Comments

                           

Thank you very much for taking the time to review this manuscript. Please find the detailed responses below and the corresponding revisions/corrections highlighted in red in the re-submitted files.

 

Point-by-point response to Comments and Suggestions for Authors

 

Comments and Suggestions for Authors

In this work, combined with machine learning models, image analysis was used to recognize differences in pepper samples after selected periods of lacto-fermentation. I think the manuscript is in the scope of the Agriculture and can be considered for the publication after minor revision. The following comments should be clarified before publish.

Comment 1: Can the quality be evaluated solely based on pictures of peppers during the fermentation period?

Response 1: As mentioned in the manuscript “The quality of processed agricultural food can be monitored using computer vision systems. These systems are important for the accurate and consistent inspection of external quality parameters [21].” [lines 65-67]

The objective has been more specified:

“The objective of this study was to distinguish and assess the changes in the flesh structure of sweet bell pepper samples after specific periods of fermentation in a non-destructive manner.” [lines 8-10]

“… this study aimed at non-destructive distinguishing bell pepper samples after specific periods of lacto-fermentation to assess the changes occurring during the process. Peppers with different fruit colors can have different properties and, consequently, processing, including fermentation, may take different courses. Therefore, pepper raw materials with completely different colors were selected. The identification of the pepper fermentation step is important and can be needed to determine after what period the greatest changes in the structure of the flesh of fermented samples occur and when the changes are the smallest. Based on this information the conditions and duration of fermentation can be established to obtain the most desired product.” [lines 92-100] 

 

It has been discussed as follows: “In our study, the application of image analysis of pepper samples and machine learning models developed based on texture image parameters allowed for the non-destructive and objective determination of differences in samples before and after selected periods of lacto-fermentation. The image texture features were used to distinguish the samples. The image texture is a function of the spatial variation of the pixel brightness intensity. Image textures can carry important information about the structure of the evaluated samples. The quantitative analyses of texture parameters from pictures provide valuable insights into product quality [28, 38]. Therefore the changes in the quality of peppers during the fermentation were evaluated based on pictures in terms of flesh structure.” [lines 332-340]

 

Comment 2: The author used multiple models to classify and recognize peppers, but the comparison of recognition results among different models should be discussed.

Response 2: The discussion has been improved and supplemented with information about the comparison of recognition results among different models as follows:

“Furthermore, the accuracy of distinguishing pepper samples depended on the machine learning algorithm used to build the models. Among the many tested algorithms from the groups of Functions, Bayes, Meta, Lazy, Trees, and Rules, the highest results were obtained for IBk (Lazy) – 93%, WiSARD (Functions) – 92%, and Random Committee (Meta) – 88% in the case of red bell pepper and LMT (Trees) – 90% and IBk (Lazy) – 89% for yellow bell pepper samples. Also, literature data reported the usefulness of image analysis and machine learning for the quality assessment of fermented pepper. In the previous study regarding red bell pepper, the use of IBk, SMO, Random Forest, Naive Bayes, Filtered Classifier, and JRip machine learning algorithms allowed for the distinguishing fresh and lacto-fermented samples for 6 months with an average accuracy of 99% [39].” [lines 349-359]

 

Comment 3: The applicability and scalability of the models should be discussed.

Response 3: It has been discussed as follows:

“The obtained results can be of practical application. A vision system that takes into account the use of imaging devices can be used to determine the correct course of the fermentation process. Additionally, knowing the course of fermentation for different cultivars, texture parameters of the flesh of pepper raw materials can be useful for planning the process. However, the applied approach also has limitations. The practical problem of fermentation is the fact that the course of the process and the quality of fermented products depend on the cultivar, degree of maturity, and physiochemical properties of the raw material [40,41]. To scale up the laboratory experiments to practical applications, it is necessary to collect more data for a larger number of cultivars, degrees of maturity, and growing seasons.

Although spontaneous fermentation is a reliable way of preserving fruit and vegetables, there is a risk of fermentation failure, undesirable and unpredictable changes in the quality properties, or insufficient inhibition of microorganisms causing product spoilage [42]. The application of spontaneous/natural fermentation has also disadvantages related to the fact that spoilage organisms can be easily introduced, fermentation strains are unknown, starting fermentation is difficult, and the results of the process are not controllable [43]. Therefore, in our study, slight differences could have occurred between individual technological repetitions, influencing the properties of the products and the accuracies of their classification. To avoid the problem of uncontrolled fermentation and ensure greater safety and higher quality of fermented products, selected starter cultures can be used to perform the process [44].” [lines 360-380]

 

Response to Comments on the Quality of English Language

Comment 1: Minor editing of English language required.

Response 1: We are grateful for this comment. The manuscript has been edited and English Language has been improved.

Author Response File: Author Response.pdf

Reviewer 3 Report

Comments and Suggestions for Authors

The manuscript describes the speed and intensity of paper fermentation from zero to 56 days after storage in jars. It could be useful for practical application in the production of fermented paper.

Authors should check their manuscript against the instructions for authors for publication in Agriculture and correct accordingly (max. 200 words in the abstract section).

It is not clear how the quality of the paper was evaluated. Also which quality parameters can be evaluated by non-destructive monitoring?

In the Materials and methods section, the following should be clarified: how many papers were used in the experiment, how many jars, how many papers per jar, what volume the jars had, how the temperature of 20°C was maintained, how many papers/jars were used per replicate.

The authors should be more specific in describing the amount of all ingredients that were put in each jar, and it is also not enough to say "all spices".

The manuscript does not include a title for the discussion.

It is not clear how many samples were used and monitored for this study and therefore the conclusion is questionable.

Author Response

Response to Reviewer 3 Comments

                           

Thank you very much for taking the time to review this manuscript. Please find the detailed responses below and the corresponding revisions/corrections highlighted in red in the re-submitted files.

 

Comment 1: The manuscript describes the speed and intensity of paper fermentation from zero to 56 days after storage in jars. It could be useful for practical application in the production of fermented paper.

Authors should check their manuscript against the instructions for authors for publication in Agriculture and correct accordingly (max. 200 words in the abstract section).

Response 1: Thank you very much for this comment. The manuscript has been checked and corrected according to the instructions for authors. The number of words in the abstract has been reduced to less than 200.

 

Comment 2: It is not clear how the quality of the paper was evaluated. Also which quality parameters can be evaluated by non-destructive monitoring?

Response 2: The changes in the image texture parameters were evaluated. Image textures are the external quality parameters, which indicate changes in the structure of the flesh. Image textures were extracted from color images. Image analysis is a non-destructive and objective technique. The selected image textures of pepper samples after specific periods of latco-fermentation were compared to determine the effect of this process on the external quality of samples.

It has been more clarified in the manuscript as follows:

“The quality of processed agricultural food can be monitored using computer vision systems. These systems are important for the accurate and consistent inspection of external quality parameters [21].” [lines 65-67]

The objective has been more specified: “The objective of this study was to distinguish and assess the changes in the flesh structure of sweet bell pepper samples after specific periods of fermentation in a non-destructive manner.” [lines 8-10]

“… this study aimed at non-destructive distinguishing bell pepper samples after specific periods of lacto-fermentation to assess the changes occurring during the process. Peppers with different fruit colors can have different properties and, consequently, processing, including fermentation, may take different courses. Therefore, pepper raw materials with completely different colors were selected. The identification of the pepper fermentation step is important and can be needed to determine after what period the greatest changes in the structure of the flesh of fermented samples occur and when the changes are the smallest. Based on this information the conditions and duration of fermentation can be established to obtain the most desired product.” [lines 92-100] 

 

The evaluation of the quality has been discussed as follows: “In our study, the application of image analysis of pepper samples and machine learning models developed based on texture image parameters allowed for the non-destructive and objective determination of differences in samples before and after selected periods of lacto-fermentation. The image texture features were used to distinguish the samples. The image texture is a function of the spatial variation of the pixel brightness intensity. Image textures can carry important information about the structure of the evaluated samples. The quantitative analyses of texture parameters from pictures provide valuable insights into product quality [28, 38]. Therefore the changes in the quality of peppers during the fermentation were evaluated based on pictures in terms of flesh structure.” [lines 332-340]

 

The features with the highest discriminative power responsible for the success of classification models have been presented in the manuscript as follows:

“The following image texture parameters of red bell pepper were characterized by the highest discriminative power: RSGArea, RS5SH1SumVarnc, RS5SH1DifEntrp, RS5SH3SumVarnc, RS5SZ3SumVarnc, RATeta1, GHPerc01, BS4RNGLevNonU, LHPerc01, XHPerc01, XSGNonZeros, XSGPerc01, XS5SH1SumVarnc, XS5SZ1AngScMom, UHPerc10, and UHDomn01.” [lines 201-206]

 

“In the case of the yellow bell pepper, the image textures with the highest discriminative power, which were the most important for building the most successful models were: RHMaxm01, RS5SV1SumVarnc, RS4RVLngREmph, GS5SZ5DifVarnc, LHSkewness, LHKurtosis, aHSkewness, aHKurtosis, aHDomn10, XHSkewness, YHMean, ZS5SZ3DifEntrp, US4RVGLevNonU, UASigma, SHPerc99, SS5SH5InvDfMom, SS5SN5SumVarnc, and SASigma.” [lines 281-286]

 

 

Comment 3: In the Materials and methods section, the following should be clarified: how many papers were used in the experiment, how many jars, how many papers per jar, what volume the jars had, how the temperature of 20°C was maintained, how many papers/jars were used per replicate.

Response 3: It has been clarified as follows:

“The pepper fruit samples without signs of disease, softening, or mechanical dam-age were washed thoroughly with water, then cut into four parts and the seeds were removed. The pepper pieces were put into sterile 1000 mL glass jars with standard spices for fermentation (garlic, mustard seeds, black pepper, bay leaves, Jamaica pepper - the amount of spices was up to 2 % of the quantity of pepper in the jar). Then the jars were filled with tap water with added salt so that the final NaCl content in the brine was 3.5%. The closed jars were kept in an air-conditioned room at 20°C to perform the spontaneous fermentation. The experiment was carried out in three technological repetitions per cultivar per one period of analysis, with three jars of fermented peppers in each repetition.” [lines 106-114]

 

Comment 4: The authors should be more specific in describing the amount of all ingredients that were put in each jar, and it is also not enough to say "all spices".

Response 4: Thank you for this valuable comment. It has been specified as follows:

“The pepper pieces were put into sterile 1000 mL glass jars with standard spices for fermentation (garlic, mustard seeds, black pepper, bay leaves, Jamaica pepper - the amount of spices was up to 2 % of the quantity of pepper in the jar). Then the jars were filled with tap water with added salt so that the final NaCl content in the brine was 3.5%.”[lines 108-111]

 

Comment 5: The manuscript does not include a title for the discussion.

Response 5: The title of subsection “4. Discussion” has been added. [line 331]

 

Comment 6: It is not clear how many samples were used and monitored for this study and therefore the conclusion is questionable.

Response 6: It has been clarified as follows:

“The image acquisition was performed in one hundred repetitions. For each pepper class, ten images with ten pieces in each image were obtained. During the image processing, each image was cropped into ten images including one pepper piece in each. Therefore in total one hundred images were obtained for each group.” [lines 139-143]

„Each dataset included 800 cases (100 cases for each of the 8 pepper classes).” [lines 167-168]

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

All comments were addressed