Page 64 - Ai V2.0 Flipbook C6
P. 64

3.  Adaptability: Human beings are able to adapt to varying lighting, angles, or
                      new surroundings easily. Computer Vision usually needs retraining or

                      calibration to work in uncharted territories.
                       For example, a human being is able to recognise a face in low light, while an

                      AI may require further data to accommodate that shift.

                                   4.  Fatigue: Humans tire and  may make errors when  overworked.  However,
                                       Computer Vision systems never get tired and offer consistent performance.

                                        For example, a security guard may overlook a detail after working for hours,
                                       while an AI-based CCTV can observe around the clock without rest.

                  5.  Learning: Human learning is shaped by experience, memory, and emotion.
                      Children, for example, learn to recognise letters with assistance from teachers

                      and through practice repeated over time. Computer Vision systems learn
                      from labelled sets of data and feedback.

                       For example, AI identifies letters in scanned documents using Optical Character Recognition
                      (OCR).

                                   6.   Accuracy: Humans can miss the internal detail or interpret it with prejudice.
                                       AI, if trained well, can be very accurate—but it might not grasp the bigger

                                       picture.

                                         For example, a doctor can miss the initial symptoms of a disease, whereas AI
                                       can recognise patterns that suggest the symptoms, but may also send up
                                       warning flags on innocuous cases.

                  7.  Decision-making: Human decisions are driven by emotion, ethics, and logic.
                      Computer Vision systems only decide based on rules and data entered.

                       For example, someone may allow a person to pass without an ID due to

                      empathy, but an AI system will block entry if the data doesn't align.
                                  8.   Real-life use case: Humans employ their brain and eyes to look at traffic

                                       lights and  choose when  to proceed  or  when  to brake.  Self-driving cars,
                                       on the other hand, employ Computer Vision via cameras and sensors to
                                       identify traffic lights and determine what to do. Both are doing pretty much

                                       the same thing, but the methods they employ are quite different.
                  Human vision is more intuitive and flexible, whereas computer vision is more accurate,

                  consistent, and fast, especially when handling repetitive or large-scale tasks. However, their
                  complementary roles in fields such as healthcare, security, and transportation can produce
                  remarkable results.




                           62     Touchpad Artificial Intelligence (Ver. 2.0)-VI
   59   60   61   62   63   64   65   66   67   68   69