Results of the summarizing process for the arXiv paper: 1605.02783v2

The measurement of biomechanical variables involved in human movement is a problem of great interest in disciplines such as occupational, ergonomic and sports science. Currently, this problem is solved through a two-step process: capturing data using impractical, intrusive and expensive devices and then using these data as input for complex models to obtain the biomechanical variables as output. This study proposes an automated, non-intrusive and cost-effective alternative to the first step by suggesting the capture of these data through images. The ultimate goal is to automate the entire calculation process of these variables. Specifically, this study focuses on estimating the discrete level of muscular load exerted by arm muscles from static images while holding a weight. The authors use Support Vector Machines for classification combined with pre-processing techniques that extract visual features using various methods such as Bag of Keypoints, Local Binary Patterns (LBP), Color Histograms and Contour Moments. The best-performing methods (LBP and Contour Moments) achieve classification performance measures (Precision, Recall, F-Measure and Accuracy) exceeding 90%. This approach represents a promising alternative to traditional methods for measuring biomechanical variables involved in human movement.

• - Measurement of biomechanical variables in human movement is important in occupational, ergonomic and sports science.
• - Current methods involve using impractical, intrusive and expensive devices to capture data.
• - This study proposes an automated, non-intrusive and cost-effective alternative using images to capture data.
• - The goal is to automate the entire calculation process of these variables.
• - Specifically, this study focuses on estimating the level of muscular load exerted by arm muscles from static images while holding a weight.
• - Support Vector Machines are used for classification combined with pre-processing techniques that extract visual features using various methods such as Bag of Keypoints, Local Binary Patterns (LBP), Color Histograms and Contour Moments.
• - The best-performing methods (LBP and Contour Moments) achieve classification performance measures exceeding 90%.
• - This approach represents a promising alternative to traditional methods for measuring biomechanical variables involved in human movement.

This study is about measuring how hard our muscles work when we hold something. It's important for jobs, sports, and making things comfortable to use. Usually, we need special devices that are expensive and can be uncomfortable. But this study found a way to use pictures instead! They used special techniques to look at the pictures and figure out how hard the muscles were working. This new way worked really well and could be a great alternative to the old way of measuring muscle work. Definitions- Biomechanical variables: measurements related to how our bodies move - Intrusive: something that invades or disturbs privacy or comfort - Cost-effective: something that is affordable and saves money - Muscular load: how much work our muscles are doing - Static images: still pictures that don't move - Support Vector Machines: a type of computer program used for classification (sorting things into groups) - Pre-processing techniques: methods used to prepare data before analyzing it - Classification performance measures: ways of measuring how well a computer program can sort things into groups

Measuring Biomechanical Variables in Human Movement: An Automated, Non-Intrusive and Cost-Effective Alternative

Biomechanical variables are of great interest to disciplines such as occupational, ergonomic and sports science. Currently, the process of measuring these variables involves a two-step approach that is both impractical and expensive. First, data must be captured using intrusive devices that can be costly. Then, complex models are used to obtain biomechanical variables from this data as output. This study proposes an automated, non-intrusive and cost-effective alternative for the first step by suggesting the capture of these data through images. The ultimate goal is to automate the entire calculation process of these variables.

The Study

This study focuses on estimating the discrete level of muscular load exerted by arm muscles from static images while holding a weight. Support Vector Machines (SVMs) were used for classification combined with pre-processing techniques that extract visual features using various methods such as Bag of Keypoints (BoK), Local Binary Patterns (LBP), Color Histograms and Contour Moments (CM). The best performing methods (LBP and CM) achieved classification performance measures exceeding 90%.

Preprocessing Techniques

The preprocessing techniques used in this study involved extracting visual features from static images using BoK, LBP, Color Histograms and CM methods. BoK extracts keypoints from an image which can then be used for feature extraction or matching between different images; it has been widely used in computer vision applications such as object recognition or tracking objects in videos. On the other hand, LBP is a texture descriptor based on local binary patterns which have been successfully applied to facial recognition tasks due to their robustness against changes in illumination conditions; they have also been applied to medical imaging tasks like segmentation or detection of lesions in mammography scans. Finally, Color Histograms represent color information contained within an image by quantifying its distribution across different color channels while Contour Moments measure shape information by computing moments about contours extracted from an image; they have been successfully applied to object recognition tasks where shape plays a major role such as recognizing handwritten characters or detecting faces in digital photographs.

Results

The results obtained show that SVMs combined with preprocessing techniques based on BoKs , LBPs , Color Histograms and CMs achieve classification performance measures exceeding 90%. These results suggest that this approach represents a promising alternative to traditional methods for measuring biomechanical variables involved in human movement since it provides an automated , non - intrusive and cost - effective solution . Furthermore , this method could potentially lead to faster calculations since all processing takes place directly on static images rather than relying on complex models .

Conclusion

In conclusion , this study demonstrates how SVMs combined with preprocessing techniques based on BoKs , LBPs , Color Histograms and CMs can provide accurate estimates of biomechanical variables involved in human movement without requiring costly equipment or complex models . This approach could potentially lead not only to faster calculations but also more accurate ones since all processing takes place directly on static images rather than relying solely on model outputs . As such , it presents a promising alternative for measuring biomechanical variables involved in human movement .