Image Recognition

 

Introduction:

A very widely accepted and used technique in today’s era is non other than Image Recognition. Before moving further in this topic let’s first see the domain where Image Recognition belongs.

Image Recognition is a wide part of Image Processing. Image processing operations can be performed in the spatial domain and frequency domain of an image. Spatial domain refers to the matrix of pixels composing an image (original pixels of the image). In this topic we are just going to focus on the Spacial domain.

What Is Image Recognition?

Image recognition, a subcategory of Computer Vision and Artificial Intelligence, represents a set of methods for detecting and analyzing images to enable the automation of a specific task. It is a technology that is capable of identifying places, people, objects and many other types of elements within an image, and drawing conclusions from them by analyzing them. To understand, it is as simple as the phrase ”What-you-see-is-what-you-get”.

How does Image Recognition works?

Let’s understand the working of Image Recognition in few following steps.

First:  As an image is actually made of “pixels”, many characteristics are extracted that image. Every single pixel is represented by a set of numbers and the range of these numbers is called the colour depth also known as bit depth. Basically the colour depth indicates the maximum number of potential colours that can be used in an image. In an (8-bit) greyscale image (black and white) each pixel has one value that ranges from 0 to 255. Most images today use 24-bit colour or higher. An RGB colour image means the colour in a pixel is the combination of red, green and blue. Each of the colours ranges from 0 to 255. This RGB colour generator shows how any colour can be generated by RGB. So a pixel contains a set of three values RGB(102, 255, 102) refers to colour #66ff66. An image 800 pixel wide, 600 pixels high has 800 x 600 = 480,000 pixels = 0.48 megapixels (“megapixel” is 1 million pixels). An image with a resolution of 1024×768 is a grid with 1,024 columns and 768 rows, which therefore contains 1,024 × 768 = 0.78 megapixels.

Second:  Once each image is converted to thousands of characteristics, label the images and put them into their respective category. This is called supervised machine learning.

Third: The huge networks in the middle can be considered as a giant filter. The images in their extracted forms enter the input side and the labels are in the output side. The purpose here is to train the networks such that an image with its features coming from the input will match the label in the right.

Forth/last: Once a model is trained, it can be used to recognize an unknown image. Important thing here to note, the new image will also go through the pixel feature extraction process.

Algorithm: Here we used the algorithm of Neural Networks Models. Convolutional Neural Networks (CNNs or ConvNets) have been widely applied in image classification, object detection or image recognition.

The typical neural networks stack the original image into a list and turn it to be the input layer. The information between neighboring pixels may not be retained. In contrast, CNNs construct the convolution layer that retains the information between neighboring pixels.

Challenges in image classification:

There are following main challenges in image classification:

1.    Intra-Class Variation

2.    Scale Variation

3.    View-Point Variation

4.    Occlusion

5.    Illumination

6.    Background Clutter

Why we need and what are the uses of Image Recognition?

Despite of challenges, considering the growing potential of computer vision, many organizations are investing in image recognition to interpret and analyze data coming primarily from visual sources for a number of uses such as medical image analysis, identifying objects in autonomous cars, face detection for security purpose, etc.

There are various tasks that image recognition can perform:

1.    Classification: It is the identification of the “class”, i.e. the category to which an image belongs. An image can have only one class. 

2.    Tagging: It is also a classification task but with a higher degree of accuracy. It can recognize the presence of several concepts or objects within an image. One or more tags can therefore be assigned to a particular image. 

3.    Detection: This is necessary when you want to locate an object in an image. Once the object is located, a bounding box is placed around the object in question.  

4.    Segmentation: This is also a detection task. Segmentation can locate an element on an image to the nearest pixel. For some cases, it is necessary to be extremely precise, as for the development of autonomous cars.

Future Scope:

1.    Assisting in the Educational System: By enabling students with learning disabilities to register knowledge in a way that is easier for them. For example, applications that rely on computer vision, allow text-to-speech options –this greatly assists visually impaired or dyslexic students to read the content.

2.    Optimizing Medical Imagery: Medical data consists of 90% of medical images, making this their largest data source in healthcare. So connecting one dot with another, these medical images will be trained by the smart image recognition technology to revolutionize the art of diagnosis – making detection of severe diseases, including cancer easier.

3.    Predicting Consumerism Behaviour: The area of brand advertisements, Ad targeting, and improving customer service was bound to benefit from the useful applications of image recognition. By targeting customer’s posted photos through IR – they can learn about their interests and consumerism behaviours.

4.    Iris Recognition Improvement: Iris recognition has been improved considerably with the help of image recognition technology that recognizes the unique patterns in the iris. One of the most important and essential applications of iris recognition is biometric identification.

Conclusion:

Through this blog we got to basic to advanced level of features present in Image Recognition. As we focused on what is Image Recognition, where we need it the most, the whole working process of Image Recognition, the challenges faces in this field but despite of challenges the importance and use is more important and tackling those challenges is to need of hour, later we discussed about the future scope.

Hence, Image recognition is an important application of AI techniques, as images usually act as sensory input for further problems to be solved.

Name: Urmisha Deodhe
IC_A_36