The objective of this assignment is to practice writing a parenthetical definition, a sentence definition, and an expanded definition of the same term. The objectives of the assignment include:
- Appreciate the importance and role of definitions in technical writing
- Understand how audience and purpose indicate the need for definition
- Differentiate between the levels of details in definition
- Select the right level of detail according to the situation
The term, object recognition, is relatively complex. It used in both neuroscience and computer science in the study of deep neural networks.
The reading situation is a graduate student explaining the term object recognition to a study participant after taking part in an eye tracking experiment.
Parenthetical definition:
The graduate student explained how object recognition (the ability to identify objects in an image or video) is tested in their eye tracking experiment to the study participant.
Sentence definition:
Object recognition is the process of identifying and categorizing objects in an image or video.
Expanded definition:
- History
Object recognition has been a goal of deep neural networks to apply artificial intelligence in daily life such as self-driving cars, face recognition, and various technology. Object recognition, the process of identifying and categorizing objects in an image or video, is better understood in brain mechanisms and researchers have identified regions of the brain, such as the perirhinal cortex, as the neurobiological basis of object recognition memory.
- Analysis of Parts
Object recognition is the process of identifying and categorizing objects in an image or video. Object recognition has been a goal of deep neural networks to apply artificial intelligence in daily life such as self-driving cars, face recognition, and various technology. Deep neural network is a type of machine learning where a computer is programmed to learn something such as object recognition. The ideas behind deep neural network are similar to the neurobiology in humans and can be analyzed in parts:
- An image is presented to the computer.
- A set of features are extracted from the image.
- The information is pooled in a manner based on learned patterns.
- Feature extraction and pooling are repeated for many layers.
- A top layer is created and the program determines the likelihood the patterns match a specific category and identifies the object.
- Visual
The following visual is a deep neural network demonstrating object recognition by identifying an image of cow. The visual below accompanies the previous analysis of parts:
Figure 1. A deep neural network demonstrating object recognition by identifying a cow.
Source: Wolfe, J. M., Kluender, K. R., Levi, D. M., Bartoshuk, L. M., Herz, R. S., Klatzky, R., & Merfeld, D. M. (2018). Sensation & Perception. Sinauer Associates is an imprint of Oxford University Press.
- Examples
Object recognition in deep neural networks have increasing impact in our daily lives. For example, deep neural networks can use object recognition to classify and diagnose skin cancer. Researchers used a dataset of clinical images to train a deep neural network and demonstrated that artificial intelligence is capable of classifying skin cancer as well as expert dermatologists.
References
Wolfe, J. M., Kluender, K. R., Levi, D. M., Bartoshuk, L. M., Herz, R. S., Klatzky, R., & Merfeld, D. M. (2018). Sensation & Perception. Sinauer Associates is an imprint of Oxford University Press.
Winters, B. D., Saksida, L. M., & Bussey, T. J. (2008). Object recognition memory: Neurobiological mechanisms of encoding, consolidation and retrieval. Neuroscience & Biobehavioral Reviews, 32(5), 1055–1070. https://doi.org/10.1016/j.neubiorev.2008.04.004
Esteva, A., Kuprel, B., Novoa, R. A., Ko, J., Swetter, S. M., Blau, H. M., & Thrun, S. (2017). Dermatologist-level classification of skin cancer with Deep Neural Networks. Nature, 542(7639), 115–118. https://doi.org/10.1038/nature21056
Leave a Reply