Point Detection System Based on Digital Image Processing and Infrared Sensors“ Final Project of Applied Bachelor of Telecommunication Engineering, Department of Electrical Engineering Semarang State Polytechnic, Semarang, under the guidance of Dr.Eni Dwi Wardihani, S.T., M.T. and Muhammad Anif, S.T., M.Eng.. The rapid development of the era will of course greatly affect all aspects of human life, including technological developments in the field of object detection, including the detection of fires. Now there are many alternative hotspot detection that can be used as a reference, for example using satellite technology and technology based on Mobile Wireless Sensor Network (MWSN). However, none of them are used by operational services for early fire detection. One way to overcome this is by applying computer vision technology in the form of object detection to detect hotspots. Therefore in this study, it is intended to detect hotspots based on digital image processing using the haar-cascade classifier method and infrared sensors. In addition, this research is expected to be able to provide an overview of the factors that can influence the success of the hotspot detection results by studying and observing the results of digital image processing using the haar-cascade classifier method. The test was carried out by implementing a system of detecting hotspots using the Raspberry Pi Cam and the MLX Sensor embedded in the Raspberry Pi for later testing. The results obtained were in the first test, with the height of a 13 cm raspberry pi camera detected 7 hotspots of the 12 hotspots detected with a maximum distance of 132 cm from the raspberry pi camera. Then at a height of 25 cm raspberry pi camera detected 5 hotspots with a maximum distance of 250 cm from the raspberry pi camera, and at a height of 40 cm raspberry pi camera detected only 1 hotspot object at a distance of 110 cm from the raspberry pi camera. Furthermore, in the second test the results obtained in the form of the height of a 13 cm raspberry pi camera, the largest system measuring the area of the hotspots are at the 4th hotspot with an average area of 3.44 cm2. Meanwhile, at a height of 25 cm raspberry pi camera, the largest system of measurement of the area of the hotspots is at the second hotspot with an average area of 2.54 cm2. At the height of the raspberry pi camera 40 cm wide obtained by 1,604 cm2 by system measurement and 1.8 cm2 by manual measurement. Then in the third test the data obtained is the farther the detection distance, the temperature reading detected by the MLX sensor and infrared thermometer will be smaller. Then in analysis calculation using haar-cascade classiffier method, the detection height of 13 cm threshold or the haar feature value obtained is 29. At the detection height of 25 cm, the threshold or haar feature value obtained is 232 and at the detection height of 40 cm the threshold obtained is 42.