A CNN methodology can yield pretty accurate results on stock prices if we look at day-to-day fluctuation in stock prices, but where this method fails is in anticipating big changes in prices that are not based on trends. In technical analysis, price patterns are used to identify transitions between rising and falling trends. A price movement pattern that may be calculated using a series of trend lines and/or curves is known as a cost sequence. Finding patterns in high-dimensional data might be difficult since it is difficult to visualize. Many different machine learning algorithms can match this high-dimensional data to predict and classify future events, but having the computer learn the match for a specific area of the dataset might be expensive. Using deep learning, this study proposes a way for identifying various stock market pricing styles. A CNN is used to find the pattern in stock market data, and projections are made based on it. The stock pattern is divided into five pieces. Price stability, stock value fall (quick decline, moderate decline), and stock value increase (rapid increase, gradual increase). The accuracy of our system is 98.17 %.