A New Method for Detecting Architectural Distortion in Mammograms by NonSubsampled Contourlet Transform and Improved PCNN

Round 1

Reviewer 1 Report

The paper presents a complete and exhaustive image processing pipeline for the evaluation of architectural distortion.

The topic is well addressed and presented in details, though the notation is too cumbersome from time to time. The approach is rather classic based on morphological operation, contourlet transforms, Otsu and PCNN though the comparison with respect to other methods shows improved specificity.

There are three main areas in which the paper should be improved:

Include a perspective in the use of tomosynthesis for architectural distortion detection and evaluation; Survey apporaches based on deep leraning (eg https://arxiv.org/abs/1807.03167) and compare the results with recpsect to them;  Rework formulas and equations in order to make the paper more readable; skip tables with image-by-image results while providing aggregated results; shorten the paper.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

There are some comments in the attached file.

Comments for author File: Comments.pdf

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 3 Report

 

 

Comments for author File: Comments.pdf

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 4 Report

 

 Upgrading&ModificationsNeeded:-

This paper is designed asvital techniquesforarchitectural distortion in breast cancerusing visual cameradata. They used severalalgorithms such as NSCT and PCNN for analysisdata well of various visual reasoningpatternscorrectly. Overall paper written well,however, need following major correctionsand modificationsvery concisely.

(In abstractsection),Authors must have focusedat uniqueness issues during abstract writing.Also, pinpoint the contributed work of their work.Please reviseagain abstract. (In introduction section), Authors need to mentioned description in different Paragraphs and ignoringrepeating similar information.Each para must has its own entity. (In introduction section), At first paragraph, authors should include morepracticalapplications of objecttracking, detection, understand as well asrecognition. For example: robotics[1]-[3], computer engineering [4], physical sciences, health-related issues, natural sciences [5] and industrial academic areas [6]. And include all belowreference.

 

[1]M. S. Bakli, M. A. Sakr and T. H. A. Soliman, “A spatiotemporal algebra in Hadoopfor moving objects,” Geo-spatial Information Science, Vol. 21, no. 2, pp. 102-114, 2018.

[2] “A Wrist Worn Acceleration Based Human Motion Analysis and Classification for Ambient Smart Home System,” Journal of Electrical Engineering & Technology, 2019.

[3] “Human body parts estimation and detection for physical sports movements,” IEEE International Conference on Communication, Computing and Digital Systems, 2019.

[4] “Students’ Behavior Mining in E-learning Environment Using Cognitive Processes with Information Technologies,” Education and Information Technologies, Springer, 2019.

[5] “Recognize facial expression using active appearance and neural network,” in Proceedings of International Conference on I-SMAC, India, Feb 2017.

[6] “Automatic facial expression recognition for affective computing based on bag of distances,” in Proceedings of Signal and Information Processing Association Annual Summit and Conference, Taiwan, pp. 1-4, Oct 2013.

(In introduction section), Authors donot need to just restrict to CADbasedtechniques. Explore further. Revise this section again. (In introduction section), Authors should add informationabout different sensors such as binary, digital cameras,depth data, and stereoscopic camerasare used in visual object detectionandtrackingfields[1-5].Also, give following referencesfor different sensorbased technology.

 

[1] Real-time vision-based camera tracking for augmented reality applications, ACM symposium for virtual reality software and technology.

[2] “A depth video-based human detection and activity recognition using multi-features and embedded hidden Markov models for health care monitoring systems, International Journal of Interactive multimedia and Artificial Intelligence, vol. 4(4), pp. 54-62, 2017..

[3] “Video summarization by clustering using euclidean distance,” in Proc. Conference on Signal Processing, Communication, Computing and Networking Technologies, 2011.

[4] “Facial Expression recognition using 1D transform features and Hidden Markov Model, Journal of Electrical Engineering & Technology, vol. 12(4), pp. 1657-1662, 2017.

[5] “Depth Images-based Human Detection, Tracking and Activity Recognition Using Spatiotemporal Features and Modified HMM, Journal of Electrical Engineering and Technology, pp. 1921-1926, 2016.

(In Introductionsection), Add new information whichmentioned about the state of the art features used in different areas such as bodytracking, action recognition, gait detection, hand tracking, human-human interaction, etc. and mentioned below references.

 

[1] Person re-identification across multi-camera system based on local descriptors,” in Proceedings IEEE conference on distributed smart cameras.

[2] “Daily Human Activity Recognition Using Depth Silhouettes and R Transformation for Smart Home,” in Proceedings Smart Homes Health Telematics, pp. 25-32, 2011.

[3] Vision-based real-time motion capture system using multiple cameras,” in Proceedings IEEE Conference on Multisensor Fusion and Integration for Intelligent Systems.

[4] “Real-Time Life Logging via a Depth Silhouette-based Human Activity Recognition System for Smart Home Services,” in Proceedings of the IEEE International Conference on Advanced Video and Signal-based Surveillance, pp. 74-80, 2014.

[5]"The Mechanism of Edge Detection using the Block Matching Criteria for the Motion Estimation," Proc. Human Computer Interaction, pp.484-489, Jan. 2005.

(In Introduction); Write down a paragraphof your system original contributions. (In Proposed Methodsection), In Fig. 1, preprocessing is not clean. Also, define sub-blocks of main blocks. (In Method of selectionsection),Formulation of Hat transformation is missing. (In Method of selectionsection), Why length is 100 and angle is 100 always?? (In Coefficient Operationsection), Please, refer different state-of-the-art worksof feature extraction, features fusion, classification and recognition Techniques. Used below latest works in these fields and mentioned below references compulsory [1-8].

 

[1] Graph formulation of video activities for abnormal activity recognition, Pattern recognition, 2017.

[3] “Salient Segmentation based Object Detection and Recognition using Hybrid Genetic Transform”, IEEE ICAEM conference, 2019.

[3] “Abnormal activity detection based on dense spatial-temporal features and one-class learning, SoICT, 2017.

[4] “Sensors Technologies for Human Activity Analysis Based on SVM Optimized by PSO Algorithm”, IEEE ICAEM conference, 2019.

[5] “Abnormal activity detection using Pyroelectric infrared sensors”, sensors, 2016.

(Processing low frequency coefficient), Design anAlgorithm for 2.1.1, 2.1.2 and 2.2, .Input and Output are unclear. (Material and Assessment method),Please mentioned datasets visual examples first. Then, explain overall descriptions.

 

(Experiments),Table 6needs properly explanations. What is experimental setup, splittingdataset strategies and computational processing time is needed? (Experiments),In Table 7, calculate performance values between state-of-the-art systems with your system. Use these state-of-the-art systems. i.e., HMM, Modified HMM, embedded HMM, GMM, SVMs, etc. [1-4]. Authors must refer following references.

 

[1]Gaussian mixture based HMM for human daily activity recognition using 3D skeleton features,” in processing inter. conf. On Industrial electronics and applications.

[2] “A depth video sensor-based life-logging human activity recognition system for elderly care in smart indoor environments,” Sensors.

[3] “Wearable Sensor-Based Human Behavior Understanding and Recognition in Daily Life for Smart Environments, IEEE conference on International Conference on Frontiers of information technology, 2018.

[4] “Human activity recognition based on the combined SVM & HMM,” in Proceedings Inter. Conf. on Information and Automation.

(Conclusion),Authors must have mentioned about failures and weak areas of their proposed system.

 

Comments for author File: Comments.pdf

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

Thank you for having addressed my comments.

Reviewer 3 Report

Authors addressed all my comments. 

Reviewer 4 Report

Revision is not upto mark. 

Answers are not satisfactory.