Challenges To Facial Recognition After Plastic Surgery
Challenges To Facial Recognition After Plastic Surgery
This paper aimed in identifying the challenges of recognizing a given face after plastic surgery; in reality it is quite a challenging task to identify human faces which have under gone some corrections with computer algorithm. Surgery in face is considered as a challenging exploration issue in the field of face recognition, apart from the challenges of face recognition after plastic surgery and feature works in this area are also highlighted to enable more exploration in this domain.
The system of verifying/identifying faces is one among the few methods in biometrics system which gets the merits of high level of accuracy and non intrusive. It provides details about age, personal identity, gender, emotional state and the mood of a person. However, even after sometime of exploration, face stands as an active topic due to the changeability perceive in the facial appearance as a result of illumination, expression occlusion, pose etc. One among the new challenges of knowing human faces is facial plastic surgery. Any face that passes through surgery alters the features to be used in knowing faces to a certain degree that human being got it hard to detect a person’s face after plastic surgery. The statistical data has shown that plastic surgery is growing (Signh, 2011). This is as because of technological increase, cheap, and the fastness in conducting the surgery, a lot of individual do undergo plastic surgery not for passion but for medical reasons and others select cosmetic surgery to look younger and better by their appearance. The method can notably change the regions of a face for what ever type of surgery; alteration is done on the features and skin texture of a face. Again, as a result of privacy issues, the surgical details of a given face are not available and any plastic surgery face database contains images before and after the face surgery. This further makes the extraction of feature difficult task in knowing human faces.
Understanding facial plastic surgery
Knowing a human face by computer is called facial recognition is the ability of a computer system to recognize a test face in a given image, from gallery or video frame (Andrew 2015). It is becoming an area for wider exploration and application especially in the field of computer vision, access control, and investigation purposes. However when a given face undergoes surgery criminals or evaders hide their identities and resides in the society smoothly, which means the standard of face recognition is compromised making it an issue to be dealt with by the use of technology. Hence to overcome these trepidations, recognition of human faces by computer had to spread its tentacles and address this issue successfully (Pankaj et al., 2018). Plastic or cosmetic surgery can be defined as the process of altering a person’s appearance through specific procedures including liposuction, rhinoplasty, facelifts, and breast implant or augmentation (Singh R. et al. 2011). Although the facial surgery is advantageous for some it can be misused by some who have indulged in some kind of crime and wants to conceal his/her identity. Many countries have incorporated facial data into the electronic passport for security reasons. However, due to the plastic surgery facial texture, shape and countenance can be altered and the security becomes unapt (Liu X. et al., 2013). In all type of plastic surgery, there is one common thing that is all are gone through some kind of facial modification and diverge from the original. Plastic surgery is broadly categorized into two (Pankaj et al., 2018). Local plastic surgery: is mainly concern with correcting a portion in a face. Global plastic surgery: concern with altering the entire structure of the face (Singh et al., 2011).
Traditionally, face recognition research has focused primarily on developing novel characterizations and algorithms to deal with challenges posed by variations in acquisition conditions like illumination conditions and head pose with respect to the camera. Tremendous success in dealing with these problems is probably one of the primary factors that has generated interest in new avenues in face matching that include matching faces across plastic surgery variations. Although the main reason why people chose facial recognition system as the best among all other biometrics system is because it is non intrusive, the accuracy of facial recognition is usually affected by substantial intra-class variations due to some factors (Lin et al., 2009), including age, pose (Jiménez et al., 2009), lighting (Takallou and Kasaei, 2014), and expression (Yi. Et al., 2014). Most current works on facial recognition focuses on the efforts of minimizing the effect of such variations that could deteriorate the overall performance of system (Singh at al., 2009). However, a preliminary study on the effect of plastic surgery on facial recognition was conducted, and they concluded that it is imperious for facial recognition systems to be able to tackle this significant matter, and that the research in this area is an imperative requirement (Singh et al., 2009). One of the first works related to recognizing faces after plastic surgery from digital images studied how different types of plastic and cosmetic surgeries can affect the performance of a number of baseline facial recognition algorithms. They concluded that available facial recognition algorithms are not able to handle plastic surgeries that are classified as global procedures including, full face lift and skin resurfacing (Singh et al., 2010), a new approach to find the nearness between the pre plastic surgical face to the post plastic surgical face. They develop a classifier for facial images that have previously undergone some feature modifications through plastic surgery based on near set theory. Their work concerned only geometrically obtained feature values and their approximation using near sets. Once the features will be extracted a feature database will be formed. Using this feature values near set theory provides a method to establish resemblance between objects contained in a disjoint set, that is it provides a formal basis for observational comparison and classification of the objects. One limitation to this approach is, it will recognize the face only after local plastic surgery, but not work in the presence of global plastic surgery (Signh 2011). A sparse representation approach combined with the popular sparse representation in 2012 was used to address the challenge of plastic surgery variations and utilizes images from sequestered non-gallery subjects with similar local facial characteristics to fulfill this requirement. But still one limitation of sparsity-based biometric recognition is, it requires several images per subject in the gallery (Gaurav et al., 2012), not until 2013 where multiobjective evolutionary approach was developed to match face images before and after plastic surgery. The algorithm first generates non-disjoint face granules at multiple levels of granularity. The first level of granularity processes the image with Gaussian and Laplacian operators to assimilate information from multiresolution image pyramids. The second level of granularity tessellates the image into horizontal and vertical face granules of varying size and information content. The third level of granularity extracts discriminating information from local facial regions. After features are extracted from that face granules by SIFT and EUCLBP algorithm. Then Multiobjective Evolutionary Approach was used to optimization of weight. Decision is take place on the basis of weight (Himanshu et al., 2013). Still in 2014 another approach was developed by Bhatt using analysis tools has not yield better result. A geometrical face recognition after plastic surgery GFRPS system is proposed the recognition process is performed in three steps; localising the regions of interest ROIs of the ‘After’ image, measuring the geometrical distances between the ROIs centres to determine the post-geometrical features vector, and using a minimum distance classifier to compare the post-features vector with the pre-features vectors database to find the perfect matching. The main advantage of the proposed system is its simplicity besides its high performance. The experimental results reveal that the proposed technique achieves much higher face identification rate than the best known results in the literature beside its high robustness under different types of plastic surgery procedures (Shaimaa and Hossam, 2014).
This work introduces a freshMultimodal Biometric approach making use of novel approaches to boostthe rate of recognition and security. The proposed method consists of various processes like Face segmentation using Active Appearance Model (AAM), Face Normalization using Kernel Density Estimate/Point Distribution Model (KDE-PDM), Feature extraction using Local Gabor XOR Patterns (LGXP) and Classification using Independent Component Analysis (ICA). Efficient techniques have been used in each phase of the FRAS in order to obtain improved results(Devi and Marimuthu, 2016 ). In 2018 a new propose method This paper proposes an approach for recognition of such faces using PCA (Principle component analysis) for dimension reduction and LBP (Local binary pattern) for feature extraction from the facial region and periocular region (Pankaj, 2018).
Recent Challenges to face recognition after plastic surgery
Facial plastic surgery changes face appearance, which intuitively affects the robustness of appearance based face recognition. In this section, we analyze the effects of different plastic surgery procedures on face appearance
Changes in skin texture:
Some plastic surgery makes people look younger or more attractive by removing face scars, acnes or taking skin resurfacing. As a result, the skin texture will change.
Changes of face component:
The main face components: forehead, eyelid, nose, lip, chin and ear can be reshaped or restructured by plastic surgery. The local skin texture around the face component may also be disturbed.
Changes of global face appearance:
Global facial plastic surgery will change the global face appearance, in other words, not only part of the face component and the skin texture will change, but also the whole face geometric structure and appearance will be disturbed. In summary, the challenges of face recognition after plastic surgery mainly lie in the fact that faces after plastic surgery have undergone various appearance changes, but no method is available to detect or model such changes.
Types of facial plastic surgery
When an individual undergoes plastic surgery, the facial features are reconstructed either globally or locally (Pankaj, 2018). Therefore, in general, plastic surgery can be classified into two distinct categories.
Disease correcting local plastic surgery (Local surgery):
This is a kind of surgery in which an individual undergoes local plastic surgery for correcting defects, anomalies, or improving skin texture. Local plastic surgery techniques can be applied for possibly three different purposes: 1) to correct by-birth anomalies, 2) to cure the defects that are result of some accident, and 3) to correct the anomalies that have developed over the years. Examples of disease correcting local plastic surgery would be surgery for correcting jaw and teeth structure, nose structure, chin, forehead and eyelids etc. Local plastic surgery is also aimed at reshaping and restructuring facial features to improve the aesthetics. This type of local surgery leads to varying amount of changes in the geometric distance between facial features but, the overall texture and appearance may look similar to the original face. However, any of the local plastic surgery procedures may be performed in conjunction with one or more such procedures and an amalgamation of such procedures may result in a fairly distinct face when compared to the original face.
Plastic surgery for reconstructing complete facial structure (Global surgery):
Apart from local surgery, plastic surgery can be performed to completely change the facial structure which is known as full face lift. Global plastic surgery is recommended for cases where functional damage has to be cured such as patients with fatal burns or trauma. Note that, global plastic surgery is primarily aimed at reconstructing the features to cure some functional damage rather than to improve the aesthetics. In this type of surgery, the appearance, texture and facial features of an individual are reconstructed to resemble normal human face but are usually not the same as the original face. Furthermore, global plastic surgery may also be used to entirely change the face appearance, skin texture and other facial geometries making it arduous for any face recognition system to recognize faces before and after surgery. Therefore, it can also be misused by criminals or individuals who want to remain elusive from law enforcement and pose a great threat to society despite all the security mechanism in-place.
In the above mentioned categories of facial plastic surgery, there are several types of surgeries which are described as follows:
Most of the existing face recognition algorithms have predominantly focused on mitigating the effects of pose, illumination and expression, while the challenges of face recognition due to aging and disguise still remains. As these procedures become more and more prevalent, future face recognition systems will be challenged to recognize individuals after plastic surgery has been performed (Richa, et.al., 2008). In all type of plastic surgery, there is one common thing that is all are gone through some kind of facial modification and diverge from the original fce; generally there are two main challenges to facial recognition after plastic surgery (Pankaj, 2018).
- Due to the sensitive nature of the process and the privacy issues involved, it is extremely difficult to prepare a face database that contains images before and after surgery (Richa, et.al., 2008).
- After surgery, the geometric relationship between the facial features changes significantly and there is no technique to detect and measure such type of alterations (Richa, et.al., 2008). And facial recognition performancedepend on facial features extracted.
There are certain features of a face which cannot be altered even after plastic surgery like the shape of the zygomatic bone, nodal points on the face and the pupillary distance (Pankaj, 2018). Now since the shape of zygomatic bone can not be change even if plastic surgery is performed on a face there is likely tendency to for feature facial recognition after plastic surgery to work with However features extracted from the face and periocular region to perform can be used to
For future directions, they recommended fusing matching scores for feature-based and texture-based algorithms using fusion techniques, also the need for robust facial recognition algorithms that can alleviate the effects of facial alteration.
A face recognition technique falters while the faces have undergone single or multiple plastic surgeries. But there are a few features which remain unaltered and not easy to be doctored. Our proposed technique is designed to use these features and successfully recognize faces which might have undergone surgeries.
In the area of recognizing faces, many techniques have been introduced with the purpose of addressing the issues of illumination, pose, expression, aging and camouflage. However plastic surgery based face recognition is still a lesser explored area. These types of surgeries changes the facial features to such large extent that often human beings find it tedious to recognize a person face once the surgery is performed. This paper introduces a novel Multimodal Biometric approach making use of a novel approaches for improving the rate of recognition and security. The proposed method consists of various processes like Face segmentation, Face Normalization, Feature extraction and Classification. In the first step for Segmenting an object out of an image is determining and defining the boundary that encapsulates that object. In several scenarios, the shape of the boundary can give a lot of data about the object itself. For the segmentation of the well-known facial features, this approach system utilizes Active Shape Models with Invariant Optimal Features (IOF- AAM). This approach integrates local image search with global shape constraints depending on a Point Distribution Model (IPDM). Then the second step introduces a Kernel Density Estimate/Point Distribution Model (KDE-PDM) algorithm that is in accordance with PCA, and the Kernel Density Estimation for providing a statistical shape model which can manage with high dimensional data sets, represent the correlated variation modes. In the third stage, feature extraction is performed to the reduction of the actual data set through the measurement of specific characteristics, or features from one to another one. The texture of the tumor is extracted using method called Local Gabor XOR Patterns (LGXP) initially forces the image to undergo a Gabor filtering, where the convolution of the image with the Gabor kernels is done to get the required output. Finally classification is done by Independent Component Analysis (ICA) for decaying a sophisticated dataset into autonomous sub-portions.