click here to download project abstract
click here to download project base paper
ABSTRACT
Identification of bird species is a challenging task often resulting in ambiguous labels. Even professional bird watchers sometimes disagree on the species given an image of a bird. It is difficult problem that pushes the limits of the visual abilities for both humans and computers. Although different bird species share the same basic set of parts, different bird species can vary dramatically in shape and appearance. Intraclass variance is high due to variation in lighting and background and extreme variation in pose (e.g., flying birds, swimming birds, and perched birds that are partially occluded by branches).Our project aims to employ the power of machine learning to help amateur bird watchers identify Bird species from the images they capture.
INTRODUCTION
Bird behavior and population trends have become an important issue now-a-day. Birds help us to detect other organisms in the environment. An important problem in ecology, which is the study of interactions between organisms and environment, is to monitor bird populations. The use of acoustics to monitor and classify birds in their natural environments has received a lot of interest lately. Classification of bird species based on image data so, for example useful when monitoring breeding behavior, biodiversity and population dynamics.
Now a day’s bird watching is a recreational activity that can provide relaxation in daily life and it’s a responsibility to know about our nature because birds are part of our society. Someone who does this is called a birdwatcher or birder. The scientific study of birds is called ornithology. People who study birds as a profession are called ornithologists. We have nearly 18,000 bird species on our beautiful earth by the new research led by American
Museum of Natural History. Birds that look similar to one another or thought to interbreed, but
actually different species.
However, because of observer constraints such as location, distance and equipment, identifying birds with naked eye is based on basic characteristics features and appropriate based on distinct features is often seen as tedious classification. Bird classification can be done manually by domain experts but growing amounts of data leads to time consuming process. Later in this detection of object parts is the challenging task because of Complex variations and fringes of objects. Our main Aim is about bird identification technology to maintain a data base of birds species like a gallery for the generations because our ancestors history data is given in the form of book
and papers but for our future generations we have to give the data by using technology So, that we built this technology for everyone to check the birds thesis easily by the image identification.
To classify the aesthetics of birds in their natural habitats, this study developed a method using
a convolutional neural network (CNN) to extract information from bird images captured
previously or in real time by identifying local features. First, raw input data myriad semantic
parts of birds were gathered and localized.
