**1. Introduction**

Brain-computer interface (BCI) technology provides a means of communication that allows individuals with severely impaired movement to communicate with assistive devices using the electroencephalogram (EEG) or other brain signals. The practicality of a BCI has been made by advances in multi-disciplinary areas of research related to neuroscience, brainimaging techniques and human-computer interfaces. The end goal of a BCI is to enable monitoring of the underlying brain processes and subsequent utilization of this information for communicating and controlling devices solely through the brain without depending on the normal output pathways of peripheral nerves and muscles. Photographs capture reality. However, this belief no longer holds true in the current digital era given that the

© 2016 The Author(s). Licensee InTech. This chapter is distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. © 2018 The Author(s). Licensee IntechOpen. This chapter is distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

manufacture of counterfeit images has increased [1]. The development of powerful photo editing software, such as Photoshop, has simplified the production of fake digital images [2]. A case of image counterfeiting is shown in **Figure 1**. Image forgery has severe consequences. For example, by modifying faces in an image, image counterfeiting can be applied to ruin a person's reputation. Academic documents may also include manipulated images that misrepresent experimental data. In addition, image forgery can be applied to remove a reference object from a standard image. As a result, the validity of the image can no longer be accepted [3]. These multilevel protection issues have different implications in different fields, such as detective work.

In simple terms, a brain-computer interface (BCI) is a direct interface between the human brain and an artificial system. Its purpose is to control the actuation of a device, say a robotic system or a wheelchair, with brain activity but without the use of peripheral nerves or muscles [4]. BCI in a literal sense means interfacing an individual's electrophysiological signals with a computer [5]. Thus, in a true sense, the BCI only uses signals from the brain and must consider eye and muscle movements as artifacts1 or noise. Information from various knowledge domains is necessary to create a complete BCI system. Thus, an artificial neural network (ANN) is an information-processing paradigm that is inspired by the way in which biological nervous systems, such as the brain, process information. This network is composed of a large number of highly interconnected processing elements referred to as neurons that work in unison to solve specific problems. Enhancing the noisy electroencephalogram (EEG) signal utilizes a layer of neurons in the spatial dimension within the neural network framework. The incoming noisy input signal sample is treated as a probability density function (pdf) by the layer of neurons and it recurrently evolves under the influence of the SWE and appropriate learning rules. This approach has made possible the development of an efficient computational algorithm referred to as the recurrent quantum neural network algorithm (RQNN) which to some extent has solved the complex problem under consideration. In general, two methods can be applied to detect image fraud: active and passive certification [6]. These two methods are illustrated in **Figure 2**. Active certification is categorized into two classes. The first class is based on the identification of a digital watermark. A watermark is hidden in the image at the end of capture, the detection program checks if the image certificate has been edited [7, 8]. The watermark is inserted when the image is taken using a specially equipped photographic camera or after acquisition by an expert [1]. The successive editing

of the original image may degrade image quality. Passive certification methods are based on digital signatures. These methods identify the distinguishing characteristics of an image as a signature after image acquisition. At the end of certification, signatures are renewed in accordance with a similar method, and the genuineness of the image can be identified by comparison. Digital signatures and watermarks have similar disadvantages. Negative image certification, also referred to as forensic digital image certification, is highly practical. Digital image certification does not require extra information and is independent of the image theme [9]. Negative methods have two parts: (1) identification of the original edit and (2) detection of tampering [10]. Certification for the first class is based on digital fingerprint certification, effects allowed by image acquisition, and storage. The methods used in this class use the digital fingerprint of the camera to differentiate among similar or dissimilar camera models. The detection methods of passive falsification can either be false or independent. Fraud detection methods are employed in particular cases of counterfeiting, similar to making copies or linking images. To discover universal forgery, researchers use autonomous techniques and exploit three different types of artifacts: the effects of resampling, pressure, and contradictions [10]. The types of counterfeiting techniques can be categorized into two classes: copy-detecting technique (image forging) and image-binding technique (two-fold

Rotation Invariant on Harris Interest Points for Exposing Image Region Duplication Forgery

http://dx.doi.org/10.5772/intechopen.76332

31

The ease and effectiveness of counterfeiting facilitates its application in changing image content [11]. The important features, like the pallet and the active range, of replicated areas are compatible with the rest of the image given that these areas are obtained from the same image [12]. Nevertheless, in practice, counterfeiting may imply more than simple replication. Numerous image-editing processes may be applied in serious counterfeiting, as shown in **Figure 3**. The processes can be divided into two groups: intermediator processes and postprocesses. Intermediator processes are applied to synchronicity and homogeneity between a replicated region and its neighbor [13]. Intermediator processes include rotation, scaling,

image-based counterfeiting).

**2. Copy-move forgery detecting**

**Figure 2.** Detect image fraud: active and passive certification.

**Figure 1.** Image forgery has severe consequences.

Rotation Invariant on Harris Interest Points for Exposing Image Region Duplication Forgery http://dx.doi.org/10.5772/intechopen.76332 31

**Figure 2.** Detect image fraud: active and passive certification.

manufacture of counterfeit images has increased [1]. The development of powerful photo editing software, such as Photoshop, has simplified the production of fake digital images [2]. A case of image counterfeiting is shown in **Figure 1**. Image forgery has severe consequences. For example, by modifying faces in an image, image counterfeiting can be applied to ruin a person's reputation. Academic documents may also include manipulated images that misrepresent experimental data. In addition, image forgery can be applied to remove a reference object from a standard image. As a result, the validity of the image can no longer be accepted [3]. These multilevel protection issues have different implications in different

In simple terms, a brain-computer interface (BCI) is a direct interface between the human brain and an artificial system. Its purpose is to control the actuation of a device, say a robotic system or a wheelchair, with brain activity but without the use of peripheral nerves or muscles [4]. BCI in a literal sense means interfacing an individual's electrophysiological signals with a computer [5]. Thus, in a true sense, the BCI only uses signals from the brain and must consider eye and muscle movements as artifacts1 or noise. Information from various knowledge domains is necessary to create a complete BCI system. Thus, an artificial neural network (ANN) is an information-processing paradigm that is inspired by the way in which biological nervous systems, such as the brain, process information. This network is composed of a large number of highly interconnected processing elements referred to as neurons that work in unison to solve specific problems. Enhancing the noisy electroencephalogram (EEG) signal utilizes a layer of neurons in the spatial dimension within the neural network framework. The incoming noisy input signal sample is treated as a probability density function (pdf) by the layer of neurons and it recurrently evolves under the influence of the SWE and appropriate learning rules. This approach has made possible the development of an efficient computational algorithm referred to as the recurrent quantum neural network algorithm (RQNN) which to some extent has solved the complex problem under consideration. In general, two methods can be applied to detect image fraud: active and passive certification [6]. These two methods are illustrated in **Figure 2**. Active certification is categorized into two classes. The first class is based on the identification of a digital watermark. A watermark is hidden in the image at the end of capture, the detection program checks if the image certificate has been edited [7, 8]. The watermark is inserted when the image is taken using a specially equipped photographic camera or after acquisition by an expert [1]. The successive editing

fields, such as detective work.

30 Evolving BCI Therapy - Engaging Brain State Dynamics

**Figure 1.** Image forgery has severe consequences.

of the original image may degrade image quality. Passive certification methods are based on digital signatures. These methods identify the distinguishing characteristics of an image as a signature after image acquisition. At the end of certification, signatures are renewed in accordance with a similar method, and the genuineness of the image can be identified by comparison. Digital signatures and watermarks have similar disadvantages. Negative image certification, also referred to as forensic digital image certification, is highly practical. Digital image certification does not require extra information and is independent of the image theme [9]. Negative methods have two parts: (1) identification of the original edit and (2) detection of tampering [10]. Certification for the first class is based on digital fingerprint certification, effects allowed by image acquisition, and storage. The methods used in this class use the digital fingerprint of the camera to differentiate among similar or dissimilar camera models. The detection methods of passive falsification can either be false or independent. Fraud detection methods are employed in particular cases of counterfeiting, similar to making copies or linking images. To discover universal forgery, researchers use autonomous techniques and exploit three different types of artifacts: the effects of resampling, pressure, and contradictions [10]. The types of counterfeiting techniques can be categorized into two classes: copy-detecting technique (image forging) and image-binding technique (two-fold image-based counterfeiting).
